Papers with hallucination
Text Generation with Text-Editing Models (2022.naacl-tutorials)
Copied to clipboard
Eric Malmi, Yue Dong, Jonathan Mallinson, Aleksandr Chuklin, Jakub Adamek, Daniil Mirylenka, Felix Stahlberg, Sebastian Krause, Shankar Kumar, Aliaksei Severyn
| Challenge: | Text-editing models are a popular alternative to seq2seq for monolingual text generation tasks such as text summarization and style transfer. |
| Approach: | They propose to use text-editing models to predict edit operations applied to the source sequence and to generate outputs word-by-word from scratch. |
| Outcome: | This paper provides an overview of the text-edit based models and their current state-of-the-art approaches. |
Understanding Figurative Meaning through Explainable Visual Entailment (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing models for visual entailment and visual question-answering have limited ability to understand figurative meaning in images and captions. |
| Approach: | They propose a task framing the figurative meaning understanding problem as an explainable visual entailment task where the model has to predict whether the image entitles a caption and justify the predicted label with a textual explanation. |
| Outcome: | The proposed dataset contains 6,027 image, caption, label, explanation instances covering five diverse figurative phenomena. |
Struc-Bench: Are Large Language Models Good at Generating Complex Structured Tabular Data? (2024.naacl-short)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have advanced capabilities but produce complex structured data. |
| Approach: | They propose a structure-aware fine-tuning method to bolster LLMs' performance by crafting format-specific instructions from the intended outputs. |
| Outcome: | The proposed method outperforms LLMs on all three formats and spans text tables, HTML, and LaTeX formats. |
CHATREPORT: Democratizing Sustainability Disclosure Analysis through LLM-based Tools (2023.emnlp-demo)
Copied to clipboard
Jingwei Ni, Julia Bingler, Chiara Colesanti-Senni, Mathias Kraus, Glen Gostlow, Tobias Schimanski, Dominik Stammbach, Saeid Ashraf Vaghefi, Qian Wang, Nicolas Webersinke, Tobias Wekhof, Tingyu Yu, Markus Leippold
| Challenge: | a lack of transparency in sustainability reporting is a key challenge due to the sheer volume and complexity of sustainability reports . only a few entities worldwide have the resources to analyze these reports at scale . a novel LLM-based system to automate the analysis of corporate sustainability reports is needed . |
| Approach: | They propose a novel LLM-based system to automate the analysis of corporate sustainability reports. |
| Outcome: | The proposed system automates the analysis of corporate sustainability reports. |
Generating Text from Language Models (2023.acl-tutorials)
Copied to clipboard
| Challenge: | a growing percentage of natural language processing tasks focus on the generation of text from probabilistic language models. |
| Approach: | They will provide a centralized discussion of critical considerations when choosing how to generate from a language model. |
| Outcome: | This tutorial will provide a centralized discussion of critical considerations when choosing how to generate from a language model. |
Controllable Neural Dialogue Summarization with Personal Named Entity Planning (2021.emnlp-main)
Copied to clipboard
| Challenge: | Experimental results show that our proposed framework generates fluent and factually consistent summaries under various planning controls using both objective metrics and human evaluations. |
| Approach: | They propose a controllable neural generation framework that can guide dialogue summarization with personal named entity planning. |
| Outcome: | The proposed framework generates fluent and factually consistent summaries under various planning controls using objective metrics and human evaluations. |
A Cause-Effect Look at Alleviating Hallucination of Knowledge-grounded Dialogue Generation (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing dialogue systems have demonstrated impressive performance conducting fluent and natural-sounding conversations, but they are plagued by the Knowledge Hallucination problem. |
| Approach: | They propose a method that exploits the dialogue-knowledge interaction to reduce hallucination by using external knowledge resources to generate more informative responses. |
| Outcome: | The proposed method reduces hallucination without disrupting other dialogue performance while keeping adaptive to different generation models. |
Equipping Language Models with Tool Use Capability for Tabular Data Analysis in Finance (2024.eacl-short)
Copied to clipboard
| Challenge: | Large language models (LLMs) have an array of reasoning capabilities but face limitations such as error propagation and hallucination. |
| Approach: | They propose to use a LLAMA-2 13B CHAT model to act as a task router and task solver to offload certain reasoning steps to external tools that are more suited for the task. |
| Outcome: | The proposed model improves by 35.2% and 5.06% over baseline models and strong GPT-3.5 results. |
A Grounded Preference Model for LLM Alignment (2024.findings-acl)
Copied to clipboard
Tahira Naseem, Guangxuan Xu, Sarathkrishna Swaminathan, Asaf Yehudai, Subhajit Chaudhury, Radu Florian, Ramón Astudillo, Asim Munawar
| Challenge: | Large Language Models (LLMs) suffer from factual inconsistency and hallucination despite recent advances . training a preference model requires substantial human annotation, which is expensive and labor-intensive. |
| Approach: | They propose to generate synthetic grounded preference data and train a Grounded Preference Model to assess the overall quality of grounded responses. |
| Outcome: | The proposed model can generate much better grounded responses as judged by GPT4 and achieves the TRUE faithfulness Benchmark. |
Tutorial Proposal: Hallucination in Large Language Models (2024.lrec-tutorials)
Copied to clipboard
| Challenge: | Grasping the intricacies of hallucination in LLMs can be daunting, especially for those new to the field. |
| Approach: | This tutorial aims to bridge the gap between the field and the field of hallucination . it will explore the key aspects of hallucinonation, including benchmarking, detection, and mitigation techniques . |
| Outcome: | This tutorial will explore the key aspects of hallucination in LLMs . it will also explore the specific constraints and shortcomings of current approaches . |
HAD: HAllucination Detection Language Models Based on a Comprehensive Hallucination Taxonomy (2026.acl-industry)
Copied to clipboard
| Challenge: | relying on large language models for information has raised concerns about reliability and accuracy of outputs. |
| Approach: | They propose a hallucination taxonomy with 11 categories for various NLG tasks and propose HAllucination Detection models which integrate hallucinism detection, span-level identification, and correction into a single inference process. |
| Outcome: | The proposed models outperform baselines on HaluEval, FactCHD, and FaithBench, confirming their robustness and versatility. |
Addressing Bias and Hallucination in Large Language Models (2024.lrec-tutorials)
Copied to clipboard
Nihar Ranjan Sahoo, Ashita Saxena, Kishan Maharaj, Arif A. Ahmad, Abhijit Mishra, Pushpak Bhattacharyya
| Challenge: | This tutorial provides a comprehensive overview of two critical aspects of Large Language Models: bias and hallucination. |
| Approach: | This tutorial provides an overview of two critical aspects of Large Language Models: bias and hallucination. |
| Outcome: | This tutorial delves into the complex dimensions of Large Language Models (LLMs) it outlines ethical considerations pertinent to their development and discusses hallucination, a prevalent issue in generative AI systems such as LLMs. |
On the Benefits of Fine-Grained Loss Truncation: A Case Study on Factuality in Summarization (2024.eacl-short)
Copied to clipboard
| Challenge: | Text summarization and simplification are among the most widely used applications of NLP, but they are prone to hallucination due to training on unaligned data. |
| Approach: | They propose a loss-truncation approach to modify the standard log loss to adaptively remove noisy examples during training to improve model performance. |
| Outcome: | The proposed approach yields a considerable number of hallucinated entities on various datasets. |
GAPO: Learning Preferential Prompt through Generative Adversarial Policy Optimization (2025.acl-long)
Copied to clipboard
Zhouhong Gu, Xingzhou Chen, Xiaoran Shi, Tao Wang, Suhang Zheng, Tianyu Li, Hongwei Feng, Yanghua Xiao
| Challenge: | Existing methods for achieving this require a limited understanding of constraints and can be hallucinating or brittle. |
| Approach: | They propose a framework that combines adversarial training dynamics with an encoder-only reward model to progressively learn and adapt to increasingly complex constraints. |
| Outcome: | Extensive experiments show that GAPO significantly outperforms existing methods like PPO, DPO, and KTO in fine-grained constraints. |
Constructing a Dataset for Hallucination Detection in Japanese Summarization with Fine-grained Faithfulness Labels (2026.eacl-srw)
Copied to clipboard
| Challenge: | Large language models (LLMs) can generate fluent text, but the quality of generated content depends on its consistency with the given input. |
| Approach: | They constructed a Japanese evaluation dataset for hallucination detection in summarization by manually annotating sentence-level faithfulness labels in LLM-generated summaries of Japanese documents. |
| Outcome: | The proposed model can detect hallucinations in Japanese documents by annotating faithfulness labels in Japanese summaries. |
OLMoTrace: Tracing Language Model Outputs Back to Trillions of Training Tokens (2025.acl-demo)
Copied to clipboard
Jiacheng Liu, Taylor Blanton, Yanai Elazar, Sewon Min, Yen-Sung Chen, Arnavi Chheda-Kothary, Huy Tran, Byron Bischoff, Eric Marsh, Michael Schmitz, Cassidy Trier, Aaron Sarnat, Jenna James, Jon Borchardt, Bailey Kuehl, Evie Yu-Yen Cheng, Karen Farley, Taira Anderson, David Albright, Carissa Schoenick, Luca Soldaini, Dirk Groeneveld, Rock Yuren Pang, Pang Wei Koh, Noah A. Smith, Sophie Lebrecht, Yejin Choi, Hannaneh Hajishirzi, Ali Farhadi, Jesse Dodge
| Challenge: | tracing language models' outputs back to training data is a problem because they are trained on text corpora with trillions of tokens . existing methods for tracers have not been scaled to work within this multi-trillion-token setting . |
| Approach: | They propose a system that traces language models' outputs verbatim back to training data . OLMOTRACE retrieves documents from the model's training data that contain exact matches . |
| Outcome: | The proposed system can find verbatim matches between LM output and training data . it can be used to explore fact checking, hallucination, and creativity of language models . |
Long-Form Speech Translation through Segmentation with Finite-State Decoding Constraints on Large Language Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | a challenge in speech translation is that plenty of spoken content is long-form, but short units are necessary for obtaining high-quality translations. |
| Approach: | They propose a large language model to split long ASR transcripts into segments that can be independently translated to maximize translation quality. |
| Outcome: | The proposed model improves the average BLEU by 2.9 points for English–German, English–Spanish, and English–Arabic TED talk translation in 9 sets. |
Evaluating Object Hallucination in Large Vision-Language Models (2023.emnlp-main)
Copied to clipboard
| Challenge: | Large vision-language models (LVLMs) suffer from object hallucinations, i.e., they tend to generate objects inconsistent with the target images in the descriptions. |
| Approach: | They propose to integrate powerful large vision-language models (LVLMs) they propose a polling-based query method to evaluate object hallucination . |
| Outcome: | The proposed model can evaluate object hallucination in a more stable and flexible way. |
LOTUS: A Leaderboard for Detailed Image Captioning from Quality to Societal Bias and User Preferences (2025.acl-industry)
Copied to clipboard
Yusuke Hirota, Boyi Li, Ryo Hachiuma, Yueh-Hua Wu, Boris Ivanovic, Marco Pavone, Yejin Choi, Yu-Chiang Frank Wang, Yuta Nakashima, Chao-Han Huck Yang
| Challenge: | Large Vision-Language Models (LVLMs) have transformed image captioning . existing evaluations lack standardized criteria and a standardized evaluation framework . |
| Approach: | They propose a leaderboard for evaluating detailed captions that addresses three main gaps in existing evaluations: lack of standardized criteria, bias-aware assessments, and user preference considerations. |
| Outcome: | The proposed model evaluates caption quality, descriptiveness, risks, and societal biases while tailoring criteria to user preferences. |
Detecting Omissions in LLM-Generated Medical Summaries (2025.emnlp-industry)
Copied to clipboard
Achir Oukelmoun, Nasredine Semmar, Gaël de Chalendar, Clement Cormi, Mariame Oukelmoun, Eric Vibert, Marc-Antoine Allard
| Challenge: | Large Language Models (LLMs) have created a number of use cases in the medical field . omissions in summaries can jeopardize the decision-making process . |
| Approach: | They propose a dataset to evaluate omissions in large-scale medical summaries . they propose 'embedKDECheck' method that uses embeddings generated by a third-party NLP model . |
| Outcome: | The proposed method is well-suited for resource-constrained environments. |
CuriousLLM: Elevating Multi-Document Question Answering with LLM-Enhanced Knowledge Graph Reasoning (2025.naacl-industry)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have achieved significant success in open-domain question answering, however, they continue to face challenges such as knowledge cutoffs and hallucinations. |
| Approach: | They propose a new mechanism that integrates a curiosity-driven reasoning mechanism into an LLM agent to generate relevant follow-up questions. |
| Outcome: | The proposed enhancement integrates a curiosity-driven reasoning mechanism into an LLM agent, enabling it to generate relevant follow-up questions, thereby guiding the information retrieval process more efficiently. |
Verifying the Steps of Deductive Reasoning Chains (2025.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models have been shown to improve the reasoning capabilities of the models. |
| Approach: | They propose to automate verification of individual reasoning steps in a logical deductive Chain-of-Thought. |
| Outcome: | The proposed method can detect unsound reasoning steps fairly well, but under-performs symbolic methods. |
Reformatted Alignment (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Current methods to improve data quality are labor-intensive or prone to factual errors caused by LLM hallucinations. |
| Approach: | They propose a method which reformats the responses of instruction data into a format that better aligns with pre-established criteria and the collated evidence. |
| Outcome: | The proposed approach minimizes human annotation, hallucination, and the difficulty in scaling, remaining orthogonal to existing alignment techniques. |
MAC-Tuning: LLM Multi-Compositional Problem Reasoning with Enhanced Knowledge Boundary Awareness (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models produce non-existing facts when faced with questions outside their parametric knowledge, which undermines their reliability. |
| Approach: | They propose a method that separates the learning of answer prediction and confidence estimation during fine-tuning on instruction data. |
| Outcome: | Experiments on multiple models and different model sizes show that the proposed method outperforms baselines by up to 25% in average precision. |
From Generating Answers to Building Explanations: Integrating Multi-Round RAG and Causal Modeling for Scientific QA (2025.naacl-industry)
Copied to clipboard
Victor Barres, Clifton James McFate, Aditya Kalyanpur, Kailash Karthik Saravanakumar, Lori Moon, Natnael Seifu, Abraham Bautista-Castillo
| Challenge: | Application of Large Language Models to complex causal question answering can be stymied by their opacity and propensity for hallucination. |
| Approach: | They propose a causal QA approach that combines iterative RAG with a formal model of causation. |
| Outcome: | The proposed approach is implemented into a Collaborative Research Assistant (Cora) and evaluated in the life sciences domain. |
HalOmi: A Manually Annotated Benchmark for Multilingual Hallucination and Omission Detection in Machine Translation (2023.emnlp-main)
Copied to clipboard
David Dale, Elena Voita, Janice Lam, Prangthip Hansanti, Christophe Ropers, Elahe Kalbassi, Cynthia Gao, Loic Barrault, Marta Costa-jussà
| Challenge: | Previously available quality assessments do not distinguish between hallucinations and omissions. |
| Approach: | They propose to annotate hallucinations and omissions in machine translation using a single language pair. |
| Outcome: | The proposed dataset covers 18 translation directions with varying resource levels and scripts. |
TracSum: A New Benchmark for Aspect-Based Summarization with Sentence-Level Traceability in Medical Domain (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing evidence-based summarization tasks require tracing source evidence to assess their accuracy. |
| Approach: | They propose a benchmark for traceable, aspect-based summarization that pairs summaries with sentence-level citations to enable users to trace back to the original context. |
| Outcome: | The proposed benchmark can be used to evaluate document summarization with LLMs and human evaluations. |
Fidelity-Enriched Contrastive Search: Reconciling the Faithfulness-Diversity Trade-Off in Text Generation (2023.emnlp-main)
Copied to clipboard
| Challenge: | Language models often generate fluent and convincing content but can lack consistency with the provided source, resulting in potential inaccuracies. |
| Approach: | They propose a new decoding method that augments the contrastive search framework with context-aware regularization terms to promote tokens that are semantically similar to the provided source while penalizing repetitiveness in the generated text. |
| Outcome: | The proposed method improves faithfulness across various language models while maintaining output diversity comparable to well-performing decoding algorithms. |
LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions (2024.eacl-long)
Copied to clipboard
| Challenge: | Large language models with instruction tuning are resource-intensive . a recent study suggests that the performance of LLMs scales proportionally with the size of the model. |
| Approach: | They propose to distill knowledge from instruction-tuned LLMs into much smaller ones . they develop a large set of 2.58M instructions based on existing and newly-generated instructions . |
| Outcome: | The proposed models are comparable to strong baselines while being much smaller in size. |
On Large Language Models’ Hallucination with Regard to Known Facts (2024.naacl-long)
Copied to clipboard
Che Jiang, Biqing Qi, Xiangyu Hong, Dayuan Fu, Yang Cheng, Fandong Meng, Mo Yu, Bowen Zhou, Jie Zhou
| Challenge: | Large language models are successful in answering factoid questions but are also prone to hallucination. |
| Approach: | They propose self-reporting to the model when faced with such limitations. |
| Outcome: | The proposed classifier can detect hallucinations with an 88% success rate and can be used to answer factoid questions with correct answer knowledge. |
Finding and Editing Multi-Modal Neurons in Pre-Trained Transformers (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to identify key neurons for interpretability of multi-modal large language models are unclear. |
| Approach: | They propose a method to identify key neurons for interpretability by multi-modal large language models. |
| Outcome: | The proposed method improves conventional works upon efficiency and applied range by removing needs of costly gradient computation. |
On the Hallucination in Simultaneous Machine Translation (2024.acl-short)
Copied to clipboard
| Challenge: | Currently, there are no studies which systematically analyze hallucination in SiMT. |
| Approach: | They conduct a comprehensive analysis of hallucination in simultaneous machine translation (SiMT) they find that halluciation is extremely severe, especially as latency increases . |
| Outcome: | The results show that it is possible to alleviate hallucination by decreasing the over usage of target-side information for SiMT. |
The Role of Exploration Modules in Small Language Models for Knowledge Graph Question Answering (2025.acl-srw)
Copied to clipboard
| Challenge: | Existing methods to integrate knowledge graphs into large language models often rely on proprietary or extremely large models . |
| Approach: | They propose to integrate knowledge graphs into reasoning processes of large language models . they propose to use simple and efficient exploration modules to handle knowledge graph traversal . |
| Outcome: | The proposed modules improve the performance of small language models on knowledge graph question answering tasks. |
HALoGEN: Fantastic LLM Hallucinations and Where to Find Them (2025.acl-long)
Copied to clipboard
| Challenge: | generative large language models produce hallucinations that are not aligned with world knowledge or input context. |
| Approach: | They propose a hallucination benchmark framework that measures hallucinism in large language models . they evaluate 150,000 generations from 14 language models and find they are riddled with hallucinos . |
| Outcome: | The proposed framework evaluates 150,000 generations from 14 language models. |
You Truly Understand What I Need : Intellectual and Friendly Dialog Agents grounding Persona and Knowledge (2022.findings-emnlp)
Copied to clipboard
Jungwoo Lim, Myunghoon Kang, Yuna Hur, Seung Won Jeong, Jinsung Kim, Yoonna Jang, Dongyub Lee, Hyesung Ji, DongHoon Shin, Seungryong Kim, Heuiseok Lim
| Challenge: | Existing models that ground knowledge and persona at the same time are limited, leading to hallucination and a passive way of using personas. |
| Approach: | They propose a conversational agent that grounds external knowledge and persona simultaneously and a retrieval augmented generation model that generates utterances with lesser hallucination and more engagingness. |
| Outcome: | The proposed agent generates the utterance with lesser hallucination and more engagingness utilizing retrieval augmented generation with knowledge-persona enhanced query. |
Mitigating Hallucinations in Multi-modal Large Language Models via Image Token Attention-Guided Decoding (2025.naacl-long)
Copied to clipboard
| Challenge: | Multi-modal large language models (MLLMs) generate plausible but incorrect content, resulting in hallucinations . recent advances in MLLM technology have demonstrated their outstanding performance in a variety of visual tasks, such as object detection. |
| Approach: | They propose a plug-and-play method which leverages MLLMs’ internal representations to mitigate hallucinations by analyzing input and output tokens. |
| Outcome: | The proposed method exploits MLLMs’ internal representations to mitigate hallucinations. |
Citation-Enhanced Generation for LLM-based Chatbots (2024.acl-long)
Copied to clipboard
| Challenge: | Existing efforts to alleviate hallucination in chatbots require additional training and data annotation. |
| Approach: | They propose a Citation-Enhanced Generation approach that uses retrieval argumentation to generate citations and a natural language inference-based citation generation module to generate content. |
| Outcome: | The proposed method outperforms state-of-the-art methods on three benchmarks. |
Learning from LLM Agents: In-Context Generative Models for Text Casing in E-Commerce Ads (2025.emnlp-industry)
Copied to clipboard
| Challenge: | Existing NER-based transformer models are expensive and lack contextual dependencies, making them less reliable when handling unseen or ad-specific terms, e.g., brand names. |
| Approach: | They propose a two-stage approach to casing correction in e-commerce ad content that leverages Chain-of-Actions to enforce content policies while accurately handling ads-specific terms. |
| Outcome: | The proposed model outperforms existing NER-based models and achieves near-LLM performance at a fraction of the cost. |
FaithDial: A Faithful Benchmark for Information-Seeking Dialogue (2022.tacl-1)
Copied to clipboard
| Challenge: | a new benchmark for hallucination-free dialogues is based on knowledge-based conversational models that generate unsupported utterances . a recent study shows that models that are trustworthy generate unverifiable or factually incorrect statements . |
| Approach: | They propose a data-centric solution to edit hallucinated responses in the Wizard of Wikipedia benchmark. |
| Outcome: | The proposed model improves on the Wizard of Wikipedia benchmark while maintaining engaging conversations. |
LVLM-Compress-Bench: Benchmarking the Broader Impact of Large Vision-Language Model Compression (2025.findings-naacl)
Copied to clipboard
Souvik Kundu, Anahita Bhiwandiwalla, Sungduk Yu, Phillip Howard, Tiep Le, Sharath Nittur Sridhar, David Cobbley, Hao Kang, Vasudev Lal
| Challenge: | LVLMs have been shown to perform well on simple uni-modal benchmarks, but their detailed study on multi-modal models is still lacking. |
| Approach: | They propose a framework to analyze the impact of compression on LVLMs on multi-modal input driven tasks. |
| Outcome: | The proposed framework analyzes the impact of compression on generative performance of large vision language models on multi-modal input driven tasks. |
Lookback Lens: Detecting and Mitigating Contextual Hallucinations in Large Language Models Using Only Attention Maps (2024.emnlp-main)
Copied to clipboard
| Challenge: | Despite the utility and impressive capabilities of large language models, their tendency to generate hallucinations presents a significant challenge in their deployment. |
| Approach: | They propose a simple hallucination detection model based on the ratio of attention weights on the context versus newly generated tokens. |
| Outcome: | The proposed model reduces the amount of hallucinations by 9.6% in a summarization task. |
Mitigating Hallucinations in Large Vision-Language Models without Performance Degradation (2026.acl-long)
Copied to clipboard
| Challenge: | Recent advances in large vision-language models produce hallucinations that compromise output reliability. |
| Approach: | They propose a dual-stage framework for mitigating hallucinations without performance degradation . they propose semantic-aware component disentanglement and interpretable parameter updates . |
| Outcome: | The proposed model reduces hallucinations by 23.4% while maintaining 97.4% of general generative capability. |
Static Models, Dynamic World: A Unified Perspective on Temporal Perception in Large Language Models (2026.findings-acl)
Copied to clipboard
Chenhao Li, Dandan Song, Changzhi Zhou, Jun Yang, Yuhang Tian, Huipeng Ma, Guangyuan Feng, Luan Zhang, Xudong Li, Ke Duan
| Challenge: | Large language models are trained on static corpora but deployed in a dynamic world . a foundational tension remains between time and the ability to understand it . |
| Approach: | They formalize temporal queries in an information-theoretic framework based on parametric reachability of temporal premises and answers. |
| Outcome: | The proposed framework formalizes temporal queries in an information-theoretic framework based on parametric reachability of temporal premises and answers . the framework induces four temporal information regimes corresponding to internal reasoning, answer recency, premise anchoring, and genuine world indeterminacy . |
DAFNet: Dynamic Auxiliary Fusion for Sequential Model Editing in Large Language Models (2024.findings-acl)
Copied to clipboard
Taolin Zhang, Qizhou Chen, Dongyang Li, Chengyu Wang, Xiaofeng He, Longtao Huang, Hui Xue’, Jun Huang
| Challenge: | Large language models (LLMs) have shown impressive results, but still suffer from hallucination, i.e., the generation of false information. |
| Approach: | They propose a task of sequential model editing that aims to rectify mistakes continuously. |
| Outcome: | The proposed method significantly outperforms baselines in single-turn and sequential editing. |
TOBUGraph: Knowledge Graph-Based Retrieval for Enhanced LLM Performance Beyond RAG (2025.emnlp-industry)
Copied to clipboard
Savini Kashmira, Jayanaka L. Dantanarayana, Joshua Brodsky, Ashish Mahendra, Yiping Kang, Krisztian Flautner, Lingjia Tang, Jason Mars
| Challenge: | Retrieval-Augmented Generation (RAG) relies on query-chunk text-to-text similarity in the embedding space for retrieval, can fail to capture deeper semantic relationships across chunks, is highly sensitive to chunking strategies, and is prone to hallucinations. |
| Approach: | They propose a graph-based retrieval framework that first constructs the knowledge graph from unstructured data dynamically and automatically. |
| Outcome: | The proposed framework outperforms multiple RAG implementations in both precision and recall, significantly enhancing user experience through improved retrieval accuracy. |
HELPD: Mitigating Hallucination of LVLMs by Hierarchical Feedback Learning with Vision-enhanced Penalty Decoding (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing work detects hallucination by directly judging whether an object exists in an image, overlooking the association between the object and semantics. |
| Approach: | They propose a framework that incorporates hallucination feedback at both object and sentence semantic levels to alleviate over 15% of hallucinism. |
| Outcome: | The proposed framework can alleviate over 15% of hallucination even with a marginal degree of training. |
Mixture-of-Minds: Multi-Agent Reinforcement Learning for Table Understanding (2026.acl-long)
Copied to clipboard
Yuhang Zhou, Mingrui Zhang, Ke Li, Mingyi Wang, Qiao Liu, Qifei Wang, Jiayi Liu, Fei Liu, Serena Li, Weiwei LI, Mingze Gao, Abhishek Kumar, Xiangjun Fan, Zhuokai Zhao, Lizhu Zhang
| Challenge: | Large language models (LLMs) have shown promise on understanding and reasoning over tables, but current approaches remain limited. |
| Approach: | They propose a multi-agent framework that decomposes table reasoning into three specialized roles: planning, coding, and answering. |
| Outcome: | The proposed framework decomposes table reasoning into three specialized roles: planning, coding, and answering. |
Towards Mitigating LLM Hallucination via Self Reflection (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models have shown promise for generative and knowledge-intensive tasks including question-answering (QA) but the practical deployment still faces challenges, notably the issue of “hallucination”, where models generate plausible-sounding but unfaithful or nonsensical information. |
| Approach: | They propose a self-reflection methodology that incorporates knowledge acquisition and answer generation to address the issue of "hallucination" they use a set of LLMs to generate a more accurate and factually accurate answer. |
| Outcome: | The proposed approach improves factuality, consistency, and entailment of the generated answers. |
Enhancing Multi-Document Summarization with Cross-Document Graph-based Information Extraction (2023.eacl-main)
Copied to clipboard
| Challenge: | Information extraction (IE) and summarization (summarization) are closely related, but both aims to abstract the most salient information into a generated text summary. |
| Approach: | They propose to use structured IE graphs to enhance the abstractive summarization task by using cross-document IE output to incorporate an alignment loss between IE nodes and their text spans to reduce inconsistencies. |
| Outcome: | The proposed model can generate summaries that are more factual while not losing abstractiveness. |
LMOD: A Large Multimodal Ophthalmology Dataset and Benchmark for Large Vision-Language Models (2025.findings-naacl)
Copied to clipboard
Zhenyue Qin, Yu Yin, Dylan Campbell, Xuansheng Wu, Ke Zou, Ninghao Liu, Yih Chung Tham, Xiuzhen Zhang, Qingyu Chen
| Challenge: | Existing benchmarks for large vision-language models (LVLMs) are limited to ophthalmology-specific applications. |
| Approach: | They introduce a large-scale multimodal ophthalmology benchmark consisting of 21,993 instances across five ocular imaging modalities and 13 state-of-the-art LVLM representatives from closed-source, open-source and medical domains. |
| Outcome: | The proposed model shows significant performance drop in ophthalmology compared to other domains. |
Zero-knowledge LLM hallucination detection and mitigation through fine-grained cross-model consistency (2025.emnlp-industry)
Copied to clipboard
| Challenge: | Existing methods for hallucination management fail to integrate both detection and mitigation without external knowledge sources. |
| Approach: | They propose a black-box framework that leverages fine-grained cross-model consistency to detect and mitigate hallucinations in LLM outputs without external knowledge sources. |
| Outcome: | The proposed framework improves hallucination detection scores by 6-39% on a FELM dataset . it achieves 9 percentage points improvement in answer accuracy on the GPQA-diamond dataset compared to existing approaches . |
LLM Agents Implement an NLG System from Scratch: Building Interpretable Rule-Based RDF-to-Text Generators (2025.emnlp-industry)
Copied to clipboard
| Challenge: | Existing neural approaches to generate RDF-to-text are limited in their implementation. |
| Approach: | They propose a framework where the model is “trained” through collaborative interactions among multiple LLM agents rather than traditional backpropagation. |
| Outcome: | The proposed framework reduces hallucinations and fluency penalties on the WebNLG and OpenDialKG datasets. |
MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in Large Language Models (LLMs) generate plausible but factually incorrect outputs, posing serious risks to patient safety and clinical decision-making. |
| Approach: | They propose a benchmark for medical hallucination detection using 10,000 question-answer pairs derived from PubMedQA. |
| Outcome: | The proposed model achieves an F1 score as low as 0.625 for detecting 'hard' category hallucinations. |
Towards Fewer Hallucinations in Knowledge-Grounded Dialogue Generation via Augmentative and Contrastive Knowledge-Dialogue (2023.acl-short)
Copied to clipboard
| Challenge: | Existing knowledge-grounded dialogue generation models face the hallucination problem . Existing models generate inappropriate knowledge and generate inconsistent responses . |
| Approach: | They propose an Augmentative and Contrastive Knowledge Dialogue Expansion Framework to enhance existing knowledge dialogue models by polarizing optimization objectives and weak knowledge generation ability. |
| Outcome: | The proposed framework expands existing training sets and smooths the optimization objective that enables models to generate ground-truth with or without gold knowledge. |
Ask, Assess, and Refine: Rectifying Factual Consistency and Hallucination in LLMs with Metric-Guided Feedback Learning (2024.eacl-long)
Copied to clipboard
| Challenge: | Recent advances in Large Language Models (LLMs) have heralded unprecedented capabilities in information seeking and text generation, but challenges remain regarding citation errors and generating information not present in the evidence (hallucination). |
| Approach: | They propose a framework to assess citation errors and hallucination using an explicit evaluation paradigm to formulate actionable natural language feedback. |
| Outcome: | The proposed approach improves correctness, fluency, and citation quality and reduces hallucinations in the results. |
Open-World Factually Consistent Question Generation (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for question generation suffer from factual inconsistencies and incorrect entities and are not answerable from the input paragraph. |
| Approach: | They propose a data processing technique based on de-lexicalization for consistent question generation across domains and a model that is generic across question-generation models. |
| Outcome: | The proposed method produces entity-level factually consistent questions without significant impact on traditional metrics. |
Knowledge Verification to Nip Hallucination in the Bud (2024.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies have shown that large language models generate responses that sound plausible but contradict factual knowledge, a phenomenon known as hallucination. |
| Approach: | They propose a novel approach to align large language models to evaluate knowledge boundaries based on external knowledge to reduce hallucinations . |
| Outcome: | The proposed approach reduces hallucinations across six benchmarks using foundation LLMs of varying backbones and scales. |
How Trustworthy are Open-Source LLMs? An Assessment under Malicious Demonstrations Shows their Vulnerabilities (2024.naacl-long)
Copied to clipboard
| Challenge: | Rapid progress in open-source Large Language Models (LLMs) is driving AI development, but lacks sufficient trustworthiness to detect and mitigate adversarial demonstrations. |
| Approach: | They propose an extended Chain of Utterances-based (CoU) prompting strategy to attack open-source LLMs. |
| Outcome: | The proposed attack strategy is based on malicious demonstrations and toxicity tests on open-source models. |
Contrastive Decoding Reduces Hallucinations in Large Multilingual Machine Translation Models (2024.eacl-long)
Copied to clipboard
| Challenge: | Hallucinations occur when the target side sentence is detached from the source side sentence, or in other words, when there is a low contribution of the source sentence to the generation of the target sentence. |
| Approach: | They propose to use Contrastive Decoding to maximise the log-likelihood difference between a model and the same model with reduced contribution from the encoder outputs. |
| Outcome: | The proposed algorithm maximises the log-likelihood difference between a model and the same model with reduced contribution from the encoder outputs. |
The Troubling Emergence of Hallucination in Large Language Models - An Extensive Definition, Quantification, and Prescriptive Remediations (2023.emnlp-main)
Copied to clipboard
Vipula Rawte, Swagata Chakraborty, Agnibh Pathak, Anubhav Sarkar, S.M Towhidul Islam Tonmoy, Aman Chadha, Amit Sheth, Amitava Das
| Challenge: | Recent advances in Large Language Models have generated widespread acclaim, but hallucination has also emerged as a by-product. |
| Approach: | They propose a fine-grained discourse on profiling hallucination based on its degree, orientation, and category . they categorize hallucines into six types: acronym ambiguity, generated golem, virtual voice, geographic erratum, time wrap . |
| Outcome: | The proposed method categorizes hallucination into six types based on their degree, orientation, and category . |
Plausible May Not Be Faithful: Probing Object Hallucination in Vision-Language Pre-training (2023.eacl-main)
Copied to clipboard
| Challenge: | Large-scale vision-language pre-trained (VLP) models generate unfaithful or nonsensical texts given the source input, which is called hallucination. |
| Approach: | They propose a VLP loss-based model to mitigate object hallucination by decoupling VLP objectives and a token-level image-text alignment. |
| Outcome: | The proposed model reduces object hallucination by 17.4% on two benchmarks. |
Training Language Models to Generate Text with Citations via Fine-grained Rewards (2024.acl-long)
Copied to clipboard
| Challenge: | Recent Large Language Models (LLMs) are prone to hallucination and their outputs often contain incorrect or unverifiable claims. |
| Approach: | They propose a training framework using fine-grained rewards to teach LLMs to generate highly supportive and relevant citations while ensuring the correctness of their responses. |
| Outcome: | The proposed training framework outperforms existing methods on QA datasets and surpasses GPT-3.5-turbo on LLaMA-2-7B. |
Training Dynamics for Text Summarization Models (2022.findings-acl)
Copied to clipboard
| Challenge: | Pre-trained language models have shown impressive results when fine-tuned on large summarization datasets. |
| Approach: | They analyze the training dynamics for generation models, focusing on summarization . they find that a propensity to copy the input is learned early in the training process . |
| Outcome: | The proposed model learns at different stages of fine-tuning, the authors show . they show that factual errors are learnt in later stages, but not at high-loss tokens . |
Neural Path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding (2021.emnlp-main)
Copied to clipboard
| Challenge: | Dialogue systems that generate factually incorrect responses are often unfitful and hallucinate factuality invalid. |
| Approach: | They propose a method to improve faithfulness and reduce hallucination of neural dialogue systems to known facts supplied by a Knowledge Graph. |
| Outcome: | The proposed approach improves faithfulness and reduces hallucination of dialogue systems to known facts . it leverages a token-level fact critic to identify plausible sources of hallucinism . |
LLM-XTM: Enhancing Cross-Lingual Topic Models with Large Language Models (2026.acl-long)
Copied to clipboard
| Challenge: | Existing cross-lingual topic models depend on sparse bilingual resources and often yield incoherent or weakly aligned topics. |
| Approach: | They propose a framework that integrates LLM-guided topic refinement with self-consistency uncertainty quantification to enable black-box, stable, and scalable enhancement of cross-lingual topic models. |
| Outcome: | Experiments on multilingual corpora show that the proposed framework achieves superior topic coherence and alignment while reducing reliance on bilingual dictionaries and expensive LLM calls. |
Mitigating Hallucinated Translations in Large Language Models with Hallucination-focused Preference Optimization (2025.naacl-long)
Copied to clipboard
| Challenge: | Machine Translation (MT) systems based on fine-tuned large language models (LLMs) are at a higher risk of generating hallucinations, which can severely undermine user’s trust and safety. |
| Approach: | They propose a method that intrinsically learns to mitigate hallucinations during the model training phase. |
| Outcome: | The proposed method reduces hallucinations by 89% on an average across three unseen target languages while preserving translation quality. |
Cracking the Code of Hallucination in LVLMs with Vision-aware Head Divergence (2025.acl-long)
Copied to clipboard
Jinghan He, Kuan Zhu, Haiyun Guo, Junfeng Fang, Zhenglin Hua, Yuheng Jia, Ming Tang, Tat-Seng Chua, Jinqiao Wang
| Challenge: | Existing methods focus on alignment training or decoding refinements but address symptoms at the generation stage without probing the underlying causes. |
| Approach: | They propose a training-free approach to mitigate hallucination by enhancing the role of vision-aware attention heads. |
| Outcome: | The proposed method achieves superior performance compared to state-of-the-art approaches in mitigating hallucinations while maintaining high efficiency with negligible additional time overhead. |
Unified Hallucination Detection for Multimodal Large Language Models (2024.acl-long)
Copied to clipboard
Xiang Chen, Chenxi Wang, Yida Xue, Ningyu Zhang, Xiaoyan Yang, Qiang Li, Yue Shen, Lei Liang, Jinjie Gu, Huajun Chen
| Challenge: | despite significant strides in multimodal tasks, MLLMs are plagued by the critical issue of hallucination. |
| Approach: | They propose a meta-evaluation benchmark to facilitate evaluation of advancements in hallucination detection methods. |
| Outcome: | The proposed framework validates hallucinations robustly and provides strategic insights . MHaluBench is a meta-evaluation benchmark designed to facilitate evaluation . |
Sources of Hallucination by Large Language Models on Inference Tasks (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are claimed to be capable of Natural Language Inference (NLI) |
| Approach: | They propose to use LLMs to probe their behavior using controlled experiments. |
| Outcome: | The proposed models perform significantly worse on NLI test samples which do not conform to these biases than those which do. |
Uncertainty Quantification for In-Context Learning of Large Language Models (2024.naacl-long)
Copied to clipboard
Chen Ling, Xujiang Zhao, Xuchao Zhang, Wei Cheng, Yanchi Liu, Yiyou Sun, Mika Oishi, Takao Osaki, Katsushi Matsuda, Jie Ji, Guangji Bai, Liang Zhao, Haifeng Chen
| Challenge: | Existing studies on in-context learning have focused on quantifying the uncertainty associated with the model's response, but they neglect the complexity of the LLM and the uniqueness of in-constitut learning. |
| Approach: | They propose a method to quantify the uncertainty associated with in-context learning and propose corresponding estimation method to quantify both types of uncertainties. |
| Outcome: | The proposed method offers an unsupervised way to understand the prediction of in-context learning in a plug-and-play fashion. |
Word Alignment as Preference for Machine Translation (2024.emnlp-main)
Copied to clipboard
| Challenge: | Hallucination and omission are a problem in machine translation because of an LLM's size and low-resource languages. |
| Approach: | They propose to use word alignment as preference to optimize an LLM-based MT model to mitigate hallucination and omission problems. |
| Outcome: | The proposed model is able to mitigate hallucination and omission by using word alignment as preference. |
Benchmarking Hallucination in Large Language Models Based on Unanswerable Math Word Problem (2024.lrec-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) are highly effective in various natural language processing tasks, but can produce unreliable conjectures in ambiguous contexts, which is known as hallucination. |
| Approach: | They propose a method to evaluate LLM hallucination in Question Answering based on the unanswerable math word problem (UMWP) . they combine text similarity and mathematical expression detection to determine whether LLM considers the question unanswered. |
| Outcome: | The proposed method combines text similarity and mathematical expression detection to determine whether the LLM considers the question unanswerable. |
DSCD: Large Language Model Detoxification with Self-Constrained Decoding (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for decoding large language models (LLMs) are based on external constraints and require additional resource overhead and loss of generation fluency. |
| Approach: | They propose a method for LLMs detoxification without parameter fine-tuning that strengthens the inner token distribution while weakening that of hallucination and toxic layer during output generation. |
| Outcome: | Extensive experiments on open-source LLMs and public datasets demonstrate DSCD's state-of-the-art (SOTA) performance in detoxification and generation fluency, with superior efficiency compared to existing methods. |
Visual Evidence Prompting Mitigates Hallucinations in Large Vision-Language Models (2025.acl-long)
Copied to clipboard
| Challenge: | LVLMs have shown impressive progress by integrating visual perception with linguistic understanding to produce contextually grounded outputs. |
| Approach: | They propose a visual evidence prompting method to mitigate hallucinations in large vision-language models by using small visual models to complement them. |
| Outcome: | The proposed method reduces hallucinations by reducing false activation and enhancing correct ones. |
Chain-of-Verification Reduces Hallucination in Large Language Models (2024.findings-acl)
Copied to clipboard
Shehzaad Dhuliawala, Mojtaba Komeili, Jing Xu, Roberta Raileanu, Xian Li, Asli Celikyilmaz, Jason Weston
| Challenge: | Large language models can generate plausible but incorrect factual information, termed hallucinations, but they can still fail on lesser known facts. |
| Approach: | They develop a method that allows language models to deliberate on the responses they give in order to correct their errors. |
| Outcome: | The proposed method decreases hallucinations across a variety of tasks, including list-based questions, closed book MultiSpanQA and longform text generation. |
Learn to Refuse: Making Large Language Models More Controllable and Reliable through Knowledge Scope Limitation and Refusal Mechanism (2024.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in large language models have demonstrated impressive language understanding and generation capabilities, enabling them to answer a wide range of questions across various domains. |
| Approach: | They propose a refusal mechanism that instructs LLMs to refuse to answer challenging questions in order to avoid errors. |
| Outcome: | The proposed approach improves the controllability and reliability of large language models and their ability to answer questions across domains. |
Improve Decoding Factuality by Token-wise Cross Layer Entropy of Large Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Large language models (LLMs) often struggle with the issue of generating inaccurate or fabricated content even when they possess correct knowledge. |
| Approach: | They propose a decoding method that mitigates hallucinations without extra training . they propose entropy eNhanced decoding that leverages inner probability changes . |
| Outcome: | The proposed method improves the truthfulness and informativeness of generation while maintaining robust QA accuracy. |
RiddleBench: A New Generative Reasoning Benchmark for LLMs (2026.findings-eacl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) show remarkable capabilities, but complex reasoning skills require deeper investigation. |
| Approach: | They propose a benchmark of 1,737 puzzles to test reasoning beyond simple pattern matching. |
| Outcome: | The proposed model performs poorly when faced with reordered constraints or irrelevant information. |
Once Correct, Still Wrong: Counterfactual Hallucination in Multilingual Vision-Language Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing hallucination benchmarks rarely test this failure mode outside Western contexts and English. |
| Approach: | They propose a multimodal benchmark built from images spanning 17 MENA countries . they use a CFHR-based test to measure hallucination beyond raw accuracy . |
| Outcome: | The proposed model is based on images from 17 MENA countries . it measures counterfactual acceptance conditioned on correctly answering the true statement. |
On Hallucination and Predictive Uncertainty in Conditional Language Generation (2021.eacl-main)
Copied to clipboard
| Challenge: | Modern deep neural network models have brought drastic improvements in generation quality measured by standard metrics on different natural language generation tasks. |
| Approach: | They propose a beam search extension to reduce hallucination in conditional language generation by adding a prediction extension to beam search. |
| Outcome: | The proposed extension improves trading performance on standard metric for less hallucination with the proposed beam search variant. |
FFE-Hallu: Hallucinations in Fixed Figurative Expressions: A Benchmark of Idioms and Proverbs in the Persian Language (2026.eacl-long)
Copied to clipboard
| Challenge: | Figurative language, especially fixed figurative expressions, poses unique challenges for large language models . Unlike literal phrases, FFEs are culturally grounded and often non-compositional, making them vulnerable to figurativ hallucination . |
| Approach: | They propose a benchmark to evaluate LLMs' ability to generate, detect, and translate fixed figurative expressions in Persian. |
| Outcome: | The proposed benchmarks show that LLMs still struggle with figurative language expressions . the benchmarks are based on 600 carefully curated examples spanning three tasks . |
Getting Sick After Seeing a Doctor? Diagnosing and Mitigating Knowledge Conflicts in Event Temporal Reasoning (2024.findings-naacl)
Copied to clipboard
| Challenge: | Event temporal reasoning aims at identifying the temporal relations between two or more events from narratives. |
| Approach: | They propose to detect knowledge conflicts in event temporal reasoning using bias indicators such as event relation prior bias, tense bias, narrative bias, and dependency bias. |
| Outcome: | The proposed method can be applied to Pre-trained Language Models and Large Language Model (LLMs) as additional training data or demonstrations for In- Context Learning. |
What Makes a Good Query? Measuring the Impact of Human-Confusing Linguistic Features on LLM Performance (2026.findings-eacl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are often treated as defects of the model or its decoding strategy. |
| Approach: | They construct a 22-dimension query feature vector covering clause complexity, lexical rarity, anaphora, negation, answerability, and intention grounding. |
| Outcome: | The proposed model covers clause complexity, lexical rarity, anaphora, negation, answerability, and intention grounding, all known to affect human comprehension. |
Do Androids Know They’re Only Dreaming of Electric Sheep? (2024.findings-acl)
Copied to clipboard
| Challenge: | Detecting hallucinations in grounded generation tasks is commonly framed as a textual entailment problem. |
| Approach: | They develop probes that are narrowly trained to predict hallucination in a transformer language model. |
| Outcome: | The probes can detect hallucinations at many transformer layers outperforming baselines and human annotators on two out of three generation tasks. |
Elevating Legal LLM Responses: Harnessing Trainable Logical Structures and Semantic Knowledge with Legal Reasoning (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing approaches to large language models focus on semantic similarity, neglecting the intricate logical structures and reasoning essential for addressing complex legal issues. |
| Approach: | They propose a Logical-Semantic Integration Model (LSIM) that bridges semantic and logical coherence and a supervised framework that integrates semantic features with in-context learning. |
| Outcome: | The proposed framework significantly improves accuracy and reliability on a real-world legal QA dataset. |
From Single to Multi: How LLMs Hallucinate in Multi-Document Summarization (2025.findings-naacl)
Copied to clipboard
| Challenge: | a recent study investigated hallucinations in multi-document summarization tasks . but, it is unclear how challenges arising from handling multiple documents affect outputs . |
| Approach: | They investigate how hallucinations manifest in large language models when summarizing topic-specific information from a set of documents. |
| Outcome: | The proposed benchmarks show that the models generate more hallucinations than baselines . the results highlight the need for more effective approaches to mitigate hallucinosity in MDS . |
PoLLMgraph: Unraveling Hallucinations in Large Language Models via State Transition Dynamics (2024.findings-naacl)
Copied to clipboard
| Challenge: | Existing studies have recognized hallucination as a notable concern in large autoregressive language models (LLMs). |
| Approach: | They propose a polygraph for large language models that detects "hallucination" they demonstrate that hallucination can be detected by tractable probabilistic models . |
| Outcome: | The proposed model outperforms state-of-the-art methods on open-source LLMs by 20% on TruthfulQA benchmarks. |
Leveraging Pretrained Models for Automatic Summarization of Doctor-Patient Conversations (2021.findings-emnlp)
Copied to clipboard
Longxiang Zhang, Renato Negrinho, Arindam Ghosh, Vasudevan Jagannathan, Hamid Reza Hassanzadeh, Thomas Schaaf, Matthew R. Gormley
| Challenge: | Using pretrained transformer models for automatically summarizing doctor-patient conversations presents challenges . limited training data, domain shift, long and noisy transcripts, and high target summary variability are challenges compared to human annotators. |
| Approach: | They propose a method for fine-tuning pretrained transformer models for automatically summarizing doctor-patient conversations directly from transcripts. |
| Outcome: | The proposed method surpasses the performance of an average human annotator and the quality of previous published work for the task. |
MCiteBench: A Multimodal Benchmark for Generating Text with Citations (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing work focuses on generating citations for text-only content . experimental results reveal MLLMs struggle to ground outputs reliably when handling multimodal input . |
| Approach: | They propose a benchmark to assess the ability of MLLMs to generate text with citations in multimodal contexts. |
| Outcome: | The proposed benchmark assesses the ability of MLLMs to generate text with citations in multimodal contexts. |
Instruction Tuning with and without Context: Behavioral Shifts and Downstream Impact (2026.eacl-long)
Copied to clipboard
Hyunji Lee, Seunghyun Yoon, Yunjae Won, Hanseok Oh, Geewook Kim, Trung Bui, Franck Dernoncourt, Elias Stengel-Eskin, Mohit Bansal, Minjoon Seo
| Challenge: | Prior work on instruction tuning datasets combined these data types without examining their distinct effects. |
| Approach: | They investigate how training LLMs with or without context affects model behavior and performance . they find that using context-augmented data as the backbone for vision-language models reduces hallucination . |
| Outcome: | The proposed training with context-augmented data reduces hallucination and improves grounding in the visual domain. |
Task-Oriented Dialog Systems for the Senegalese Wolof Language (2025.coling-main)
Copied to clipboard
| Challenge: | Low-resource languages such as African ones are underrepresented in large language models limiting their performance in these languages. |
| Approach: | They propose a chatbot generation engine based on the Rasa framework and a method for projecting annotations onto the Wolof language using an in-house machine translation system. |
| Outcome: | The proposed approach performs similarly to the one obtained for French, which is a resource-rich language. |
SG-FSM: A Self-Guiding Zero-Shot Prompting Paradigm for Multi-Hop Question Answering Based on Finite State Machine (2025.findings-naacl)
Copied to clipboard
Xiaochen Wang, Junqing He, Liang Chen, Gholamreza Haffari, Yiru Wang, Zhe Yang, Xiangdi Meng, Kunhao Pan, Zhifang Sui
| Challenge: | Multi-hop Question Answering (MHQA) is a challenging task that requires models to answer multiple questions with multiple passages. |
| Approach: | They propose a self-guided prompting finite state machine to improve multi-hop reasoning abilities by iterating over multiple questions and correcting itself to improve accuracy. |
| Outcome: | The proposed approach outperforms baselines on Musique and other datasets. |
Where Did That Come From? Sentence-Level Error-Tolerant Attribution (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing task definitions exclude unsupported or hallucinated content leaving them unattributed . authors propose a new definition for sentence-level error-tolerant attribution . |
| Approach: | They propose a new definition for sentence-level error-tolerant attribution that extends attribution to include incorrect or hallucinated content. |
| Outcome: | The proposed approach reduces annotation time and facilitates hallucination fixing. |
Beyond Under-Alignment: Atomic Preference Enhanced Factuality Tuning for Large Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing work evaluates the factuality of large language models on in-domain (ID) datasets and the factuality on out-of-domain datasets. |
| Approach: | They propose a framework that enhances model’s awareness of factuality at the granularity of individual facts and propose 'Atomic Preference Enhanced Factuality Tuning' this framework enhances the model’ s awareness and accuracy of factual information at the level of individual factual facts. |
| Outcome: | The proposed framework improves model performance by an average of on ID and OOD datasets, which is highly effective. |
Understanding New-Knowledge-Induced Factual Hallucinations in LLMs: Analysis and Interpretation (2026.findings-acl)
Copied to clipboard
| Challenge: | Prior studies have shown that fine-tuning on new knowledge can induce factual hallucinations in large language models (LLMs), leading to incorrect outputs when evaluated on previously known information. |
| Approach: | They propose to conduct a fine-grained analysis of large language models using a dataset Biography-Reasoning and QA and knowledge reasoning tasks to understand their findings. |
| Outcome: | The proposed model is able to perform a range of downstream tasks without requiring a large amount of knowledge and is compared with a control dataset. |
InfoGain-RAG: Boosting Retrieval-Augmented Generation through Document Information Gain-based Reranking and Filtering (2025.emnlp-main)
Copied to clipboard
Zihan Wang, Zihan Liang, Zhou Shao, Yufei Ma, Huangyu Dai, Ben Chen, Lingtao Mao, Chenyi Lei, Yuqing Ding, Han Li
| Challenge: | Retrieval-Augmented Generation (RAG) frameworks struggle with identifying whether retrieved documents meaningfully contribute to answer generation. |
| Approach: | They propose a document-related metric to quantify the contribution of retrieved documents to correct answer generation. |
| Outcome: | The proposed framework outperforms existing approaches on both single and multiple retrieval paradigms. |
An Audit on the Perspectives and Challenges of Hallucinations in NLP (2024.emnlp-main)
Copied to clipboard
Pranav Narayanan Venkit, Tatiana Chakravorti, Vipul Gupta, Heidi Biggs, Mukund Srinath, Koustava Goswami, Sarah Rajtmajer, Shomir Wilson
| Challenge: | 103 peer-reviewed publications on hallucination in large language models (LLMs) are characterized by a lack of agreement with the term ‘hallucination’ in the field of NLP. |
| Approach: | They examine 103 peer-reviewed publications on hallucination in large language models (LLMs) and conduct a survey with 171 practitioners from the field of NLP and AI to capture varying perspectives on halllucination. |
| Outcome: | The findings highlight the need for explicit definitions and frameworks outlining hallucination within NLP and highlight potential challenges. |
The Reasoning Trap: How Enhancing LLM Reasoning Amplifies Tool Hallucination (2026.acl-long)
Copied to clipboard
| Challenge: | Recent studies suggest that strengthening reasoning often coincides with increased hallucination . however, no prior work has examined whether reasoning enhancement itself causes tool hallucinism . |
| Approach: | They propose a diagnostic benchmark measuring tool hallucination in two failure modes . they demonstrate a causal relationship between enhancing reasoning and tool hallubulation . |
| Outcome: | The proposed benchmark measures tool hallucination in two failure modes: no tool available, and (ii) only distractor tools available. |
Right for Right Reasons: Large Language Models for Verifiable Commonsense Knowledge Graph Question Answering (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing Knowledge Graph Question Answering (KGQA) methods focus on answering factual questions, leaving questions involving commonsense reasoning unaddressed. |
| Approach: | They propose a commonsense KGQA methodology that axiomatically surfaces commonsensical knowledge of Large Language Models and grounding every factual reasoning step on KG triples. |
| Outcome: | The proposed method outperforms existing methods and reduces instances of hallucination and reasoning errors. |
On the Origin of Hallucinations in Conversational Models: Is it the Datasets or the Models? (2022.naacl-main)
Copied to clipboard
| Challenge: | Existing knowledge-grounded conversational benchmarks produce factually invalid statements, a phenomenon commonly called hallucination. |
| Approach: | They conduct a human study on knowledge-grounded conversational benchmarks and state-of-the-art models. |
| Outcome: | The findings raise important questions on the quality of existing datasets and models. |
Enabling Large Language Models to Generate Text with Citations (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing work relies on commercial search engines and human evaluation, making it difficult to reproduce and compare different modeling approaches. |
| Approach: | They propose a new generation paradigm that requires large language models to provide citations to one or a few text passages for any statement they generate. |
| Outcome: | The proposed model improves factual correctness and verifiability of large language models by providing citations to a set of questions and retrieval corpora and generating answers with citation. |
Visual Attention Reasoning via Hierarchical Search and Self-Verification (2026.acl-long)
Copied to clipboard
| Challenge: | Multimodal Large Language Models (MLLMs) often hallucinate due to fragile, linear reasoning and weak visual grounding. |
| Approach: | They propose a framework that reformulates reasoning as a hierarchical search with self-verification and replaces linear Chain-of-Thought with a tree-search policy capable of backtracking to correct logical errors. |
| Outcome: | The proposed framework outperforms state-of-the-art methods on hallucination and safety benchmarks. |
MAF: Multi-Aspect Feedback for Improving Reasoning in Large Language Models (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to enhance Language Models fail to address diverse error types . generic feedback is a bottleneck for addressing diverse errors in reasoning chains . |
| Approach: | They propose an iterative refinement framework that integrates multiple feedback modules . they propose to address errors in reasoning chains by integrating frozen LMs with external tools . |
| Outcome: | The proposed framework improves performance in Mathematical Reasoning and Logical Entailment by 20% and 18% respectively. |
Investigating Memorization of Conspiracy Theories in Text Generation (2021.findings-acl)
Copied to clipboard
| Challenge: | Existing studies examine conspiracy theories in social media, but they have not evaluated their presence in generative language models. |
| Approach: | They examine the ability of generative language models to generate conspiracy theory text . they highlight the difficulties of this task and discuss the drawbacks . |
| Outcome: | The proposed model can generate conspiracy theories without access to training data. |
Token Knowledge: A New Perspective For Knowledge in Large Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Predicting the presence and absence of certain knowledge in large language models could aid hallucination avoidance. |
| Approach: | They propose a token knowledge dataset construction method and use the intermediate states during inference to train probes. |
| Outcome: | The proposed method increases the model's latent potential by 60% to 90% with strong out-of-distribution generalization by training on just a few dozen prompts. |
Detecting Hallucination and Coverage Errors in Retrieval Augmented Generation for Controversial Topics (2024.lrec-main)
Copied to clipboard
Tyler A. Chang, Katrin Tomanek, Jessica Hoffmann, Nithum Thain, Erin MacMurray van Liemt, Kathleen Meier-Hellstern, Lucas Dixon
| Challenge: | a growing audience of users is engaging with LLM-driven chatbots. |
| Approach: | They propose a strategy to handle controversial topics in LLM-based chatbots based on Wikipedia’s Neutral Point of View principle. |
| Outcome: | The proposed methods detect errors in the tuned LLM responses even when no training data is available. |
Diagnosing and Mitigating Sycophancy and Skepticism in LLM Causal Judgment (2026.findings-acl)
Copied to clipboard
| Challenge: | Current evaluations obscure the answer to causal judgment in frontier models. |
| Approach: | They introduce a process-integrity evaluator that checks whether a model's answer is entailed by its own derivation, internally consistent, and not dominated by user hints under pressure. |
| Outcome: | The proposed model fails to distinguish between the two pathologies. |
On A Scale From 1 to 5: Quantifying Hallucination in Faithfulness Evaluation (2025.findings-naacl)
Copied to clipboard
| Challenge: | Hallucination is a popular topic in natural language generation (NLG). |
| Approach: | They propose to use large language models to evaluate faithfulness of guided NLGs by a rubric template and large language inference models to score the generation on quantifiable scales. |
| Outcome: | The proposed system can provide accurate judgement and explain whether a source and generation are factually consistent. |
Object Hallucination in Image Captioning (D18-1)
Copied to clipboard
| Challenge: | Existing image captioning metrics do not capture image relevance . current metrics only measure similarity to ground truth captions . |
| Approach: | They propose a new image relevance metric to evaluate captioning models with veridical visual labels and assess their rate of object hallucination. |
| Outcome: | The proposed metrics show that models with veridical visual labels have higher hallucination rates than models with lower hallucinosity. |
Entropy Guided Extrapolative Decoding to Improve Factuality in Large Language Models (2025.coling-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) exhibit impressive natural language capabilities but suffer from hallucination – generating content that does not align with realworld facts. |
| Approach: | They propose to extrapolate critical token probabilities beyond the last layer to improve decoding by manipulating the predicted distributions at inference time. |
| Outcome: | The proposed methods surpass state-of-the-art on multiple datasets by large margins. |
DAMRO: Dive into the Attention Mechanism of LVLM to Reduce Object Hallucination (2024.emnlp-main)
Copied to clipboard
| Challenge: | Despite the success of Large Vision-Language Models, they suffer from hallucination. |
| Approach: | They propose a training-free strategy that "D**ive into" the attention of LVLMs to "R**educe" object hallucination by using classification tokens of ViT. |
| Outcome: | The proposed method reduces the impact of outlier tokens on LVLMs . the proposed method is based on LLaVA-1.5, LLvaVA-NeXT and InstructBLIP . |
ANAH: Analytical Annotation of Hallucinations in Large Language Models (2024.acl-long)
Copied to clipboard
| Challenge: | a comprehensive and fine-grained measurement of the hallucination is crucial for LLMs' wide applications. |
| Approach: | They propose a dataset that offers ANalytical Annotation of Hallucinations in Large Language Models. |
| Outcome: | The proposed dataset can be used to train and evaluate hallucination annotators. |
Alleviating Hallucinations of Large Language Models through Induced Hallucinations (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing studies have shown that large language models generate inaccurate or fabricated information, a phenomenon known as hallucinations. |
| Approach: | They propose a simple strategy to induce-then-contrast decode LLMs to enhance their factuality . they first induce hallucinations from the original model and penalize them . |
| Outcome: | The proposed strategy improves factuality of large language models across task formats, model sizes, and model families. |
Does the Generator Mind Its Contexts? An Analysis of Generative Model Faithfulness under Context Transfer (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing studies have focused on examining hallucinations stemming from static input, such as in summarization or machine translation. |
| Approach: | They propose a knowledge-augmented generator that produces information that remains grounded in contextual knowledge regardless of alterations in the context. |
| Outcome: | The proposed method is designed to produce information that remains grounded in contextual knowledge, regardless of alterations in the context. |
Towards Verifiable Text Generation with Evolving Memory and Self-Reflection (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) often produce factually incorrect information, also known as hallucination. |
| Approach: | They propose a framework for verifiable text generation with evolving memory and self-reflection that incorporates long-term memory to retain documents and recent documents. |
| Outcome: | The proposed framework outperforms baselines on five datasets across three knowledge-intensive tasks. |
OCR-Memory: Optical Context Retrieval for Long-Horizon Agent Memory (2026.acl-long)
Copied to clipboard
| Challenge: | Existing LLMs are limited by text-context budgets, resulting in token-expensive storage of raw trajectories . Optical Context Retrieval Memory (OCR-Memory) renders historical tra-jectorios into images annotated with unique visual identifiers. |
| Approach: | They propose a framework that leverages the visual modality as a high-density representation of agent experience. |
| Outcome: | Optical Context Retrieval Memory (OCRM) renders historical trajectories into images annotated with unique visual identifiers. |
Hallucination Diversity-Aware Active Learning for Text Summarization (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing methods for alleviating hallucinations require costly human annotations . Existing approaches focus on a specific type of hallucinism, which limits their effectiveness . |
| Approach: | They propose a method to detect hallucinations from errors in semantic frame, discourse and content verifiability in LLM summarization using HAllucination Diversity-Aware Sampling. |
| Outcome: | The proposed framework reduces the need for costly human annotations to correct hallucinations in LLM outputs. |
Think&Cite: Improving Attributed Text Generation with Self-Guided Tree Search and Progress Reward Modeling (2025.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) are prone to hallucinations and producing factually incorrect information. |
| Approach: | They propose a framework that allows LLMs to generate citations that provide evidence for any statement. |
| Outcome: | The proposed framework outperforms baseline approaches on three datasets and significantly outperformed baseline approaches. |
RoleBreak: Character Hallucination as a Jailbreak Attack in Role-Playing Systems (2025.coling-main)
Copied to clipboard
| Challenge: | Existing approaches to combat character hallucination are vulnerable to attack . large language models (LLMs) are capable of generating responses inconsistent with intended personas . |
| Approach: | They propose a novel defence strategy that generates supplemental context through narration to mitigate role-query conflicts and improve query generalization. |
| Outcome: | The proposed defence strategy outperforms refusal-based strategies in character hallucinations and query generalization. |
CortexDebate: Debating Sparsely and Equally for Multi-Agent Debate (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to improve the reasoning performance of LLMs suffer from two major shortcomings: too lengthy input contexts and overconfidence dilemma. |
| Approach: | They propose a method to debating among LLM agents using a sparse debator graph . they use a module called McKinsey-based Debate Matter to optimize the debators . |
| Outcome: | The proposed method has been well demonstrated across eight datasets from four task types. |
Explainable Depression Detection in Clinical Interviews with Personalized Retrieval-Augmented Generation (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing systems rely on black-box neural networks, which lack interpretability, which is crucial in mental health contexts. |
| Approach: | They propose a Retrieval-augmented generation framework for Explainable depression detection that retrieves evidence from clinical interview transcripts, providing explanations for predictions. |
| Outcome: | The proposed framework retrieves evidence from clinical interview transcripts, providing explanations for predictions. |
Bold Claims or Self-Doubt? Factuality Hallucination Type Detection via Belief State (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing studies focus on detecting the presence of hallucinations but lack a systematic classification approach, which hinders deeper exploration of their characteristics. |
| Approach: | They propose a method to categorize hallucinations into two types: Overconfident and Unaware . |
| Outcome: | The proposed method categorizes factuality hallucination into two types: Overconfident and Unaware Hallucinations. |
DiaHalu: A Dialogue-level Hallucination Evaluation Benchmark for Large Language Models (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing benchmarks for hallucination detection are intentionally generated by large language models (LLMs) however, many focus on factuality while ignoring faithfulness. |
| Approach: | They propose a dialogue-level hallucination evaluation benchmark for large language models . they integrate the topic into prompts and facilitate a dialog between two LLMs . |
| Outcome: | The proposed benchmark covers four common multi-turn dialogue domains and five hallucination subtypes, extended from factuality and faithfulness hallucines. |
From Verbatim to Gist: Distilling Pyramidal Multimodal Memory via Semantic Information Bottleneck for Long-Horizon Video Agents (2026.acl-long)
Copied to clipboard
| Challenge: | Existing multimodal large language models struggle with long-horizon video understanding due to limited context windows and static memory mechanisms that fail to mirror human cognitive efficiency. |
| Approach: | They propose a pyramidal multimodal memory architecture grounded in Fuzzy-Trace Theory that structures memory hierarchically into a *Sensory Buffer*, *Episodic Stream*, and *Symbolic Schema*. |
| Outcome: | The proposed architecture achieves state-of-the-art on both offline and streaming tasks, demonstrating robust generalization and validating the effectiveness of cognition-inspired memory organization. |
Enhancing Noise Robustness of Retrieval-Augmented Language Models with Adaptive Adversarial Training (2024.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) exhibit substantial capabilities yet face challenges such as hallucination, outdated knowledge, and untraceable reasoning processes. |
| Approach: | They propose a retrieval-augmented generation approach that leverages adaptive adversarial training to dynamically adjust the model’s training process in response to retrieval noises. |
| Outcome: | The proposed approach improves the performance of the LLaMA-2 7B model under diverse noise conditions. |
Predicting Text Preference Via Structured Comparative Reasoning (2024.acl-long)
Copied to clipboard
Jing Nathan Yan, Tianqi Liu, Justin Chiu, Jiaming Shen, Zhen Qin, Yue Yu, Charumathi Lakshmanan, Yair Kurzion, Alexander Rush, Jialu Liu, Michael Bendersky
| Challenge: | Existing approaches to comparative reasoning rely on pretraining or fine-tuning models at the cost of massive human annotation and computation. |
| Approach: | They propose a model that prompts LLMs to generate structured intermediate comparisons by proposing aspects for comparison, followed by generating textual comparisons under each aspect. |
| Outcome: | The proposed model significantly reduces hallucination and improves consistency across various NLP tasks. |
Improving Factuality with Explicit Working Memory (2025.acl-long)
Copied to clipboard
Mingda Chen, Yang Li, Karthik Padthe, Rulin Shao, Alicia Yi Sun, Luke Zettlemoyer, Gargi Ghosh, Wen-tau Yih
| Challenge: | Large language models can generate factually inaccurate content, a problem known as hallucination. |
| Approach: | They propose an approach that integrates a working memory that receives feedback from external resources. |
| Outcome: | The proposed method outperforms baselines on four fact-seeking datasets and increases the factuality metric by 2 to 6 points absolute. |
OWSM-CTC: An Open Encoder-Only Speech Foundation Model for Speech Recognition, Translation, and Language Identification (2024.acl-long)
Copied to clipboard
| Challenge: | Autoregressive models can be slower during inference and have potential risks of hallucination. |
| Approach: | They propose an encoder-only speech foundation model based on Connectionist Temporal Classification. |
| Outcome: | The proposed model improves on 180k hours of public audio data for multilingual speech recognition, speech translation, and language identification. |
Beyond Noise: Characterizing Creative Potential in Unverifiable LLM Hallucinations (2026.acl-long)
Copied to clipboard
Yu Yan, Chunhong Zhang, Haiyu Zhao, Ziyang Zeng, Zihao Liu, Yongkang Wu, Jianzhou Diao, YiJie Chen, Shujie Wang, Zheng Hu
| Challenge: | Large Language Models generate outputs that extend beyond established knowledge . prior work does not characterize the unverifiable space as a whole . |
| Approach: | They propose a novelty-verifiability characterization that distinguishes Creative Synthesis from Groundless Fabrication by a conceptual creation task. |
| Outcome: | The proposed model distinguishes Creative Synthesis (Region A) from Groundless Fabrication (Regium B) it shows that Region A is non-negligible and robust, persisting across generation strategies, models, domains, and embedding choices. |
Navigating Hallucinations for Reasoning of Unintentional Activities (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing models of intentionality recognition struggle to understand the reasoning behind unintentional actions. |
| Approach: | They propose a novel prompting technique which allows the model to navigate through hallucinated thoughts to achieve better reasoning. |
| Outcome: | The proposed prompting technique outperforms standard prompting while minimizing hallucinations. |
Learning Preference Model for LLMs via Automatic Preference Data Generation (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing training methods for large language models rely on human-annotated data. |
| Approach: | They propose to learn the preference model for LLMs via automatic preference data generation (AutoPM) using HHH-guided preference data, they show reliability and potential . |
| Outcome: | The proposed approach enables LLMs to learn human preferences and align with human values. |
Steering LVLMs via Sparse Autoencoder for Hallucination Mitigation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing approaches to address hallucinations in large vision-language models require substantial computational cost and time. |
| Approach: | They propose to leverage sparse autoencoders to identify semantic directions closely associated with faithfulness or hallucination, extracting more precise and disentangled hallucinian-related representations. |
| Outcome: | The proposed method outperforms existing decoding approaches while maintaining transferability across different model architectures with negligible additional time overhead. |
Evaluating Text-to-Speech Synthesis from a Large Discrete Token-based Speech Language Model (2024.lrec-main)
Copied to clipboard
| Challenge: | Recent advances in generative language modeling applied to discrete speech tokens presented a new avenue for text-to-speech (TTS) synthesis. |
| Approach: | They propose to use generative language modeling to generate text-to-speech (TTS) outputs by a discrete token-based model. |
| Outcome: | The proposed model is rated higher in naturalness and context appropriateness in listening tests compared to a conventional TTS. |
CoDA: Restoring Contextual Dominance via Copy-Encouraged Attention Intervention for Mitigating RAG Hallucinations (2026.findings-acl)
Copied to clipboard
| Challenge: | Retrieval-augmented generation reduces hallucination by grounding outputs in external evidence. |
| Approach: | They propose a lightweight inference-time attention intervention that amplifies evidence-aligned value states to enhance contextual faithfulness and reduce hallucination. |
| Outcome: | The proposed model reduces hallucination by grounding model outputs in external evidence. |
Mathematical Proof as a Litmus Test: Revealing Failure Modes of Advanced Large Reasoning Models (2026.acl-long)
Copied to clipboard
| Challenge: | Large reasoning models have demonstrated remarkable mathematical problem-solving abilities, but their true reasoning shortcomings are often hidden. |
| Approach: | They propose to leverage the rigor and methodological complexity of mathematical proofs as a diagnostic tool to expose hidden failures. |
| Outcome: | The proposed model evaluation exploits the rigor and complexity of proof problems to uncover 10 fine-grained errors. |
RAGTruth: A Hallucination Corpus for Developing Trustworthy Retrieval-Augmented Language Models (2024.acl-long)
Copied to clipboard
| Challenge: | Retrieval-augmented generation (RAG) is a main technique for alleviating hallucinations in large language models. |
| Approach: | They propose to integrate RAG into large language models to analyze word-level hallucinations using a corpus of 18,000 naturally generated responses from diverse LLMs. |
| Outcome: | The proposed model can fine tune a relatively small LLM and achieve a competitive hallucination detection performance when compared to the existing prompt-based approaches. |
The Dawn After the Dark: An Empirical Study on Factuality Hallucination in Large Language Models (2024.acl-long)
Copied to clipboard
| Challenge: | a growing number of researchers are studying the hallucination issue in large language models. |
| Approach: | They propose a hallucination detection benchmark and a method to detect hallucines in LLMs. |
| Outcome: | The proposed method detects hallucinations and mitigates them using different training stages. |
An Empirical Study on Parameter-Efficient Fine-Tuning for MultiModal Large Language Models (2024.findings-acl)
Copied to clipboard
| Challenge: | Multimodal Large Language Models fine-tuned with multimodal instruction-following data have demonstrated formidable capabilities in multimodal tasks. |
| Approach: | They propose to employ four PEFT methods to fine-tune the LLM component of open-source MLLMs. |
| Outcome: | The proposed method is the best performing on seven datasets, while fine-tuning the connector layers leads to improved performance in most MLLMs. |
SkewRoute: Training-Free LLM Routing for Knowledge Graph Retrieval-Augmented Generation via Score Skewness of Retrieved Context (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models incur high inference costs during deployment, causing hallucination . no dedicated routing methods exist for RAG, and existing training-based routers face challenges scaling to this domain . |
| Approach: | They propose a plug-and-play routing framework that optimizes performance and cost . the framework delivers over 3x higher routing effectiveness while reducing runtime to less than 0.001x . |
| Outcome: | The proposed framework delivers over 3x higher routing effectiveness while reducing runtime to less than 0.001x compared to existing methods. |
Can Knowledge Graphs Make Large Language Models More Trustworthy? An Empirical Study Over Open-ended Question Answering (2025.acl-long)
Copied to clipboard
| Challenge: | Existing benchmarks for integrating Knowledge Graphs with Large Language Models focus on closed-ended tasks, leaving a gap in evaluating performance on more complex, real-world scenarios. |
| Approach: | They propose a benchmark to evaluate LLMs augmented with KGs in open-ended, real-world question answering settings. |
| Outcome: | The proposed benchmark reflects practical complexities through diverse question types and incorporates metrics to quantify both hallucination rates and reasoning improvements in LLM+KG models. |
What if...?: Thinking Counterfactual Keywords Helps to Mitigate Hallucination in Large Multi-modal Models (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to reduce hallucination in large multi-modal models are lacking in addressing this problem. |
| Approach: | They propose a method that implants counterfactual thinking into Large Multi-modal Models using self-generated counterfact keywords into the models. |
| Outcome: | The proposed method improves the reliability of large multi-modal models in addressing hallucination. |
Less is More: Mitigating Multimodal Hallucination from an EOS Decision Perspective (2024.acl-long)
Copied to clipboard
| Challenge: | Extensive research has shed light on the origins of multimodal hallucinations, including the inability of vision encoders to represent finegrained visual details and model reliance on inherent parametric knowledge such as language priors and statistical biases. |
| Approach: | They propose to use EOS to terminate generation of large multimodal models by comparing the generated text with the image to mitigate multimodal hallucinations. |
| Outcome: | The proposed method significantly improves the hallucination performance of Large Multimodal Models without additional data or knowledge. |
Spotlight and Shadow: Attention-Guided Dual-Anchor Introspective Decoding for MLLM Hallucination Mitigation (2026.findings-acl)
Copied to clipboard
| Challenge: | Multimodal Large Language Models (MLLMs) excel in tasks ranging from image captioning to complex reasoning. |
| Approach: | They propose a contrastive decoding framework that dynamically calibrates each token generation by mining the model’s internal perceptual discrepancies. |
| Outcome: | The proposed framework mitigates hallucination while enhancing general reasoning capabilities. |
A Benchmark and Robustness Study of In-Context-Learning with Large Language Models in Music Entity Detection (2025.coling-main)
Copied to clipboard
| Challenge: | Recent approaches to detecting music entities have been limited due to ambiguity in user-generated content. |
| Approach: | They propose to use user-generated metadata to benchmark and test models for entity detection . they find that large language models outperform SLMs in a variety of downstream tasks . |
| Outcome: | The proposed model outperforms existing models in a variety of downstream tasks . the proposed model is more robust in the ICL setting than the existing models . |
FACTCHECKMATE: Preemptively Detecting and Mitigating Hallucinations in LMs (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Language models (LMs) hallucinate. |
| Approach: | They introduce a classifier that predicts whether LMs hallucinate based on model’s hidden states before decoding begins. |
| Outcome: | The proposed model preemptively detects hallucinations by learning a classifier that predicts whether the LM will hallucinate . if a hallucinomy is detected, FactCheckmate intervenes by adjusting the model’s hidden states to produce more factual outputs. |
Exploring In-Context Learning for Knowledge Grounded Dialog Generation (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing knowledge grounded dialog generation models are prone to hallucination and produce factually inaccurate outputs. |
| Approach: | They propose a retrieval-based framework which leverages in-context learning and retrieval techniques to enhance LLMs on knowledge grounded dialog generation. |
| Outcome: | The proposed framework outperforms existing training-based models on a large-scale knowledge graph with 1M+ facts and is expected to perform knowledge-intensive tasks. |
Analyzing LLM Behavior in Dialogue Summarization: Unveiling Circumstantial Hallucination Trends (2024.acl-long)
Copied to clipboard
| Challenge: | Recent advances in large language models have improved summarization, but they still face a challenge of hallucination. |
| Approach: | They propose a taxonomy of errors to address the problem of hallucination in LLMs . they propose two prompt-based approaches for fine-grained error detection . |
| Outcome: | The proposed model outperforms existing metrics in identifying the novel "Contextual Inference" error type. |
The Impact of Negated Text on Hallucination with Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies on hallucination in large language models (LLMs) have been actively progressing in natural language processing. |
| Approach: | They propose to examine whether LLMs can recognize contextual shifts caused by negation and still reliably distinguish hallucinations comparable to affirmative cases. |
| Outcome: | The proposed model can detect hallucinations comparable to affirmative cases, but it is difficult to detect them in negated text, the authors show . |
CaPE: Contrastive Parameter Ensembling for Reducing Hallucination in Abstractive Summarization (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing work suggests that the degree of hallucination depends on factual errors in training data. |
| Approach: | They propose a method to use training data to reduce hallucination by ensembling parameter variations in training data. |
| Outcome: | The proposed method improves on XSUM and CNN/DM datasets on human evaluations and factual metrics. |
A Comprehensive Survey of Hallucination in Large Language, Image, Video and Audio Foundation Models (2024.findings-emnlp)
Copied to clipboard
| Challenge: | a growing need to understand and alleviate FMs' propensity to produce hallucinated outputs, especially in high-stakes applications. |
| Approach: | They propose a framework for detecting and mitigating hallucination in FMs . they synthesize recent advancements in detection and mitigation techniques . |
| Outcome: | The proposed framework provides valuable insights for researchers, developers, and practitioners. |
FaithLens: Detecting and Explaining Faithfulness Hallucination (2026.findings-acl)
Copied to clipboard
Shuzheng Si, Qingyi Wang, Haozhe Zhao, Yuzhuo Bai, Guanqiao Chen, Kangyang Luo, Gang Chen, Fanchao Qi, Minjia Zhang, Baobao Chang, Maosong Sun
| Challenge: | Recent progress in large language models (LLMs) has revolutionized text generation. |
| Approach: | They propose a faithfulness hallucination detection model that can provide binary predictions and corresponding explanations to improve trustworthiness. |
| Outcome: | The proposed model outperforms advanced models on 12 diverse tasks. |
MedScore: Generalizable Factuality Evaluation of Open-ended Long-form Medical Answers by Domain-adapted Claim Decomposition and Verification (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing factuality evaluation pipelines are poor matches for medical domains . existing methods are limited to objective, entity-centric, formulaic texts . |
| Approach: | They propose a pipeline to decompose medical answers into condition-aware valid facts . they use a decomposition-then-verify approach to evaluate generated text . |
| Outcome: | The proposed method extracts up to three times as many valid facts as existing methods . the resulting factuality score substantially varies by decomposition method, corpus, and used backbone LLM . |
Emma-X: An Embodied Multimodal Action Model with Grounded Chain of Thought and Look-ahead Spatial Reasoning (2025.acl-long)
Copied to clipboard
| Challenge: | Visual-Language-Action models lack the ability to generate actionable policies tailored to specific robotic embodiments. |
| Approach: | They propose an embodied multimodal action model with Grounded Chain of Thought and Look-ahead Spatial Reasoning that enhances spatial reasoning and task planning. |
| Outcome: | The proposed model improves on existing baselines in tasks requiring spatial reasoning and grounding reasoning. |
Counterfactual Debating with Preset Stances for Hallucination Elimination of LLMs (2025.coling-main)
Copied to clipboard
| Challenge: | Existing solutions to alleviate hallucination have considered utilizing LLMs’ inherent reasoning abilities to alleviating hallucinism, such as self-correction and diverse sampling methods. |
| Approach: | They propose a counterfactual multi-agent debate framework that predetermines LLMs' stances to override their inherent biases for answer inspection. |
| Outcome: | Extensive experiments on four datasets of three tasks demonstrate the superiority of the proposed framework over existing methods. |
MURMUR: Modular Multi-Step Reasoning for Semi-Structured Data-to-Text Generation (2023.findings-acl)
Copied to clipboard
| Challenge: | MURMUR generates highly faithful and correct reasoning paths that lead to 26% more logically consistent summaries on LogicNLG compared to direct prompting. |
| Approach: | They propose a neuro-symbolic modular approach to text generation from semi-structured data with multi-step reasoning that generates reasoning paths using neural and symbolic modules with specific linguistic and logical skills. |
| Outcome: | The proposed method improves on two data-to-text generation tasks, while achieving comparable performance to fine-tuned GPT-2 on out-of-domain data. |
Halwasa: Quantify and Analyze Hallucinations in Large Language Models: Arabic as a Case Study (2024.lrec-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) generate text that is factually incorrect, nonsensical, or misleading. |
| Approach: | They create a large Arabic dataset that contains 10K of LLM generated sentences and annotate it for factuality and correctness. |
| Outcome: | The proposed dataset analyzes 10K of generated sentences and finds 25% of them are factually incorrect. |
Detecting Errors through Ensembling Prompts (DEEP): An End-to-End LLM Framework for Detecting Factual Errors (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for detecting factual errors in text summarization are inadequate for the task. |
| Approach: | They propose an end-to-end large language model framework for detecting factual errors in text summarization. |
| Outcome: | The proposed framework achieves state-of-the-art (SOTA) balanced accuracy on the AggreFact-XSUM FTSOTA, TofuEval Summary-Level, and HaluEVAL Summarization benchmarks. |
Large Language Models Are Involuntary Truth-Tellers: Exploiting Fallacy Failure for Jailbreak Attacks (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing research has shown that large language models have difficulty discerning the veracity of their intrinsic answers. |
| Approach: | They propose a jailbreak attack method that generates an aligned language model for malicious output. |
| Outcome: | The proposed method achieves competitive performance with more harmful outputs. |
Null-Shot Prompting: Rethinking Prompting Large Language Models With Hallucination (2024.emnlp-main)
Copied to clipboard
| Challenge: | a series of investigations into an interesting phenomenon where performance increases in large language models when providing a prompt that causes and exploits hallucination. |
| Approach: | They propose a null-shot prompting approach that intentionally instructs LLMs to look at and utilize information from a nil section. |
| Outcome: | The proposed approach causes and exploits hallucination in large language models on a range of tasks including arithmetic reasoning, commonsense reasoning, and reading comprehension. |
LLM Agents for Education: Advances and Applications (2025.findings-emnlp)
Copied to clipboard
Zhendong Chu, Shen Wang, Jian Xie, Tinghui Zhu, Yibo Yan, Jingheng Ye, Aoxiao Zhong, Xuming Hu, Jing Liang, Philip S. Yu, Qingsong Wen
| Challenge: | Large Language Model (LLM) agents are transforming education by automating complex tasks and enhancing both teaching and learning processes. |
| Approach: | This survey analyzes recent advances in applying Large Language Model agents to educational settings . it highlights ethical issues, hallucination and overreliance, and integration with existing ecosystems . |
| Outcome: | The authors analyze the technologies enabling LLM agents and highlight key challenges in deploying them in educational settings. |
V-DPO: Mitigating Hallucination in Large Vision Language Models via Vision-Guided Direct Preference Optimization (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing large vision-language models suffer from hallucination due to over-reliance on the Large Language Model (LLM) backbone. |
| Approach: | They propose a method to improve visual context learning by using a large-scale preference learning algorithm to improve hallucination. |
| Outcome: | The proposed method improves on human-annotated hallucination datasets. |
Trust Me, I’m Wrong: LLMs Hallucinate with Certainty Despite Knowing the Answer (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Prior work on large language model (LLM) hallucinations associated with model uncertainty or inaccurate knowledge. |
| Approach: | They define and investigate a type of hallucination where a model can answer a question correctly but a perturbation causes it to produce a hallucinous response with high certainty. |
| Outcome: | The proposed mitigations outperform existing methods on CHOKE hallucinations . the findings highlight the need to understand their origins and improve mitigation strategies . |
Error Typing for Smarter Rewards: Improving Process Reward Models with Error-Aware Hierarchical Supervision (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are prone to hallucination, especially during multihop tasks. |
| Approach: | They propose a hierarchical, erroraware discriminative PRM that classifies math errors at each step and combines finegrained signals to estimate step correctness. |
| Outcome: | The proposed model outperforms the prior best in a new stateof-theart PRMScore of 67.7 on a 400Ksample dataset . |
FreshLLMs: Refreshing Large Language Models with Search Engine Augmentation (2024.findings-acl)
Copied to clipboard
Tu Vu, Mohit Iyyer, Xuezhi Wang, Noah Constant, Jerry Wei, Jason Wei, Chris Tar, Yun-Hsuan Sung, Denny Zhou, Quoc Le, Thang Luong
| Challenge: | Modern large language models often "hallucinate" plausible but factually incorrect information, which reduces their trustworthiness especially in settings where accurate and up-to-date information is critical. |
| Approach: | They develop a human evaluation procedure to measure correctness and hallucination and use it to benchmark both closed and open-source LLMs. |
| Outcome: | The proposed method outperforms both competing search engine-augmented prompting methods and commercial systems on search-augmented QA. |
Transforming Sequence Tagging Into A Seq2Seq Task (2022.emnlp-main)
Copied to clipboard
| Challenge: | Pretrained, large, generative language models have had great success in a wide range of sequence tagging and structured prediction tasks. |
| Approach: | They propose to use a new format for casting input text sentences and their output labels into the input and target of a Seq2Seq model and introduce it to test their hypothesis. |
| Outcome: | The proposed format shows to be both simpler and more effective and devoid of hallucination. |
Chain-of-Note: Enhancing Robustness in Retrieval-Augmented Language Models (2024.emnlp-main)
Copied to clipboard
| Challenge: | Standard RALMs often neglect their intrinsic knowledge due to the interference from retrieved information. |
| Approach: | They propose a new approach to improve robustness of RALMs by generating sequential reading notes for each retrieved document. |
| Outcome: | The proposed approach outperforms standard RALMs on four open-domain QA benchmarks. |
Improving Faithfulness by Augmenting Negative Summaries from Fake Documents (2022.emnlp-main)
Copied to clipboard
| Challenge: | Current abstractive summarization systems tend to hallucinate unfaithful content . however, the most common method does not disentangle factual errors from other errors. |
| Approach: | They propose a back-translation-style approach to augment negative samples that mimic factual errors made by the model. |
| Outcome: | The proposed method improves faithfulness without sacrificing informativeness . it incorporates negative samples into training, and produces faithful/unfaithful summaries . |
Mitigating Hallucination in Fictional Character Role-Play (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Influence of parametric knowledge of large language models (LLMs) often causes role-playing characters to act out of character and hallucinate about things outside the scope of their knowledge. |
| Approach: | They propose a method that modulates the influence of parametric knowledge using a pre-calibrated confidence threshold to mitigate hallucination in fictional character role-play. |
| Outcome: | The proposed method reduces the factual accuracy of generated responses by 18% for adversarial questions and 44% in temporal hallucination for time-sensitive interviews. |
CRUSH4SQL: Collective Retrieval Using Schema Hallucination For Text2SQL (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing Text-to-SQL generators require the entire schema to be encoded with the user text. |
| Approach: | They propose a method that uses an LLM to hallucinate a minimal DB schema . they use the hallucinated schema to retrieve a subset of the actual schema based on multiple dense retrievals . |
| Outcome: | The proposed method leads to significantly higher recall than existing methods. |
Hallucination Detection in Structured Query Generation via LLM Self-Debating (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Hallucination remains a key challenge in applying large language models to structured query generation . we propose the Self-Debating framework to enhance detection performance . |
| Approach: | They propose a framework that prompts an LLM to generate contrastive explanations from opposing perspectives . they also propose 'self-debating' framework to enhance detection performance . |
| Outcome: | The proposed framework outperforms LLM-as-a-Judge baselines in hallucination detection . the framework generates contrastive explanations from opposing perspectives . |
Simple Factuality Probes Detect Hallucinations in Long-Form Natural Language Generation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Current approaches to detect hallucination require many samples from the LLM generator . current methods require multiple samples, which is computationally infeasible . |
| Approach: | They propose a simple baseline for detecting hallucinations in long-form LLM generations . they show that LLM hidden states are highly predictive of factuality in long form natural language generation . |
| Outcome: | The proposed method is comparable to expensive multi-sample approaches while drawing only a single sample from the LLM generator. |
Lost in Diffusion: Uncovering Hallucination Patterns and Failure Modes in Diffusion Large Language Models (2026.findings-acl)
Copied to clipboard
| Challenge: | dLLMs have emerged as a promising non-autoregressive paradigm for text generation, but their hallucination mechanisms remain underexplored. |
| Approach: | They present the first controlled comparative study to evaluate hallucination patterns in Diffusion Large Language Models. |
| Outcome: | The proposed model exhibits higher propensity for hallucination than AR counterparts controlled for architecture, scale, and pre-training weights. |
Tokenization Consistency Matters for Generative Models on Extractive NLP Tasks (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Pretrained sequence-to-sequence (seq2sequ) models have been widely used to solve extractive tasks, where parts of the input are extracted to form the desired output. |
| Approach: | They propose a simple fix to tokenization inconsistency that damages extractive nature of generative models by causing performance drop and hallucination. |
| Outcome: | The proposed model performs better in both in-domain and out-of-domain datasets with a notable average of +1.7 F1 gain when a BART model is trained on SQuAD and evaluated on 8 QA datasets. |
HiGoE: Hierarchical Graph of Evidence to Enhance Retrieval-Augmented Generation for Long-context Summarization (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for long-context summarization fail to capture high-level thematic structures and long-range dependencies. |
| Approach: | They propose a hierarchical Graph of Evidence to reduce hallucination and attention dilution by replacing unreliable chunk-based methods with a filtered proposition–evidence graph. |
| Outcome: | Experiments show that HiGoE surpasses baselines in quality and efficiency. |
Vision-Flan: Scaling Human-Labeled Tasks in Visual Instruction Tuning (2024.findings-acl)
Copied to clipboard
Zhiyang Xu, Chao Feng, Rulin Shao, Trevor Ashby, Ying Shen, Di Jin, Yu Cheng, Qifan Wang, Lifu Huang
| Challenge: | Recent vision-language models (VLMs) have shown impressive capabilities as general visual assistants, but there are two challenges to their performance: (1) lacking task diversity in pretraining and visual instruction tuning; (2) annotation error and bias in GPT-4 synthesized instruction tuning data. |
| Approach: | They propose a two-stage instruction tuning framework that fine tunes VLMs firstly and further tuned on GPT-4 synthesized data. |
| Outcome: | The proposed framework outperforms the traditional single-stage visual instruction tuning framework and achieves state-of-the-art performance across a wide range of multi-modal evaluation benchmarks. |
Why LLMs Hallucinate on Structured Knowledge: A Mechanistic Analysis of Reasoning over Linearized Representations (2026.acl-long)
Copied to clipboard
Shanghao Li, Jinda Han, Yibo Wang, Yuanjie Zhu, Zihe Song, Langzhou He, Kenan Kamel A Alghythee, Philip S. Yu
| Challenge: | Existing literature primarily addresses this problem through external interventions such as retrieval augmentation and prompt engineering at the input or output level. |
| Approach: | They find that LLMs can still produce hallucinated outputs when using structured external knowledge. |
| Outcome: | The proposed models fail to ground the provided knowledge, causing the model to revert to parametric memory. |
Improbable Bigrams Expose Vulnerabilities of Incomplete Tokens in Byte-Level Tokenizers (2025.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies have revealed that tokenizers can be exploited to elicit unwanted behavior. |
| Approach: | They propose to exploit incomplete tokens with stray bytes to exploit their dependency . they propose to use improbable bigrams to exploit the dependency of their adjacent tokens . |
| Outcome: | The proposed tokenizers can be exploited to elicit unwanted behavior in language models. |
CToolEval: A Chinese Benchmark for LLM-Powered Agent Evaluation in Real-World API Interactions (2024.findings-acl)
Copied to clipboard
| Challenge: | a benchmark is designed to evaluate the capabilities of large language models (LLMs) as agents in decision making and operational tasks. |
| Approach: | They propose a benchmark to evaluate LLMs in the context of Chinese societal applications . they propose he benchmark will evaluate tool invocation ability of LLM and task completion ability . |
| Outcome: | The proposed benchmark features 398 APIs across 27 widely-used Apps across 14 domains. |
Presentations are not always linear! GNN meets LLM for Text Document-to-Presentation Transformation with Attribution (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing approaches to generate presentations from document to slide are difficult to implement and cause hallucination. |
| Approach: | They propose a graph-based solution that uses a combination of graph neural network and LLM to generate a presentation with attribution of content for each slide. |
| Outcome: | The proposed approach is more efficient than using LLMs for generating a presentation from the text of a document. |
Navigating Noisy Feedback: Enhancing Reinforcement Learning with Error-Prone Language Models (2024.findings-emnlp)
Copied to clipboard
Muhan Lin, Shuyang Shi, Yue Guo, Behdad Chalaki, Vaishnav Tadiparthi, Ehsan Moradi Pari, Simon Stepputtis, Joseph Campbell, Katia Sycara
| Challenge: | Reward hacking is a problem in reinforcement learning where the ability to specify the desired behavior of a reward function is difficult. |
| Approach: | They propose to use feedback as a potential-based shaping function to solicit and apply feedback from large language models to improve convergence speed and policy returns. |
| Outcome: | The proposed method improves convergence speed and policy returns over baselines even with significant ranking errors and eliminates the need for complex post-processing of reward functions. |
FFN-SkipLLM: A Hidden Gem for Autoregressive Decoding with Adaptive Feed Forward Skipping (2024.emnlp-main)
Copied to clipboard
| Challenge: | Autoregressive Large Language Models (LLMs) are omnipresent but typically come with a substantial model size. |
| Approach: | They propose a novel fine-grained skip strategy for autoregressive large language models . they observe the saturation of computationally expensive feed-forward blocks of LLMs . |
| Outcome: | The proposed method can skip 25-30% of FFN blocks with marginal change in performance on knowledge-intensive generation tasks. |
Is the Answer in the Text? Challenging ChatGPT with Evidence Retrieval from Instructive Text (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Generative language models have recently shown remarkable success in generating answers to questions in a given textual context, but they suffer from hallucination, wrongly cite evidence, and spread misleading information. |
| Approach: | They propose a benchmark to evaluate an annotated WikiHow article and use it to retrieve answers to questions from trustworthy texts. |
| Outcome: | The proposed model can retrieve answers to lexically varied and open-ended questions from trustworthy instructive texts. |
FactAlign: Long-form Factuality Alignment of Large Language Models (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models have demonstrated significant potential as the next-generation information access engines, but reliability is hindered by issues of hallucination and generating non-factual content. |
| Approach: | They propose a novel alignment framework that enhances the factuality of LLMs’ long-form responses while maintaining their helpfulness. |
| Outcome: | The proposed framework improves factuality of LLMs while maintaining helpfulness. |
Conservative Bias in Large Language Models: Measuring Relation Predictions (2025.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) exhibit pronounced conservative bias in relation extraction tasks, often defaulting to no_relation label when an appropriate option is unavailable. |
| Approach: | They systematically evaluate the trade-off between conservative bias and hallucination in relation extraction tasks by using SBERT and LLM prompts to quantify this effect. |
| Outcome: | The proposed model defaults to no_relation label twice as often as hallucination, resulting in significant information loss when reasoning is not explicitly included in the output. |
Unleashing the Native Recommendation Potential: LLM-Based Generative Recommendation via Structured Term Identifiers (2026.findings-acl)
Copied to clipboard
Zhiyang Zhang, Junda She, Kuo Cai, Bo Chen, Shiyao Wang, Xinchen Luo, Qiang Luo, Ruiming Tang, Han Li, Kun Gai, Guorui Zhou
| Challenge: | Existing methods for constructing item identifiers face bottlenecks due to their large output space and expensive vocabulary expansion and alignment training. |
| Approach: | They propose to use Large Language Models to develop general-purpose, semantically-aware recommender systems that can be generalized and reusable. |
| Outcome: | Experiments on real-world datasets show that GRAM outperforms baselines and significantly outperformed baselines. |
From Descriptive Richness to Bias: Unveiling the Dark Side of Generative Image Caption Enrichment (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) have enhanced the capacity of vision-language models to caption visual text. |
| Approach: | They compare standard-format captions and recent GCE processes from the perspectives of gender bias and hallucination. |
| Outcome: | The proposed methods amplify gender bias by 30.9% and increase hallucination by 59.5%. |
Treble Counterfactual VLMs: A Causal Approach to Hallucination (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing studies link hallucination to data or representation biases, but their causal origins remain unclear. |
| Approach: | They propose a causal framework to analyze and mitigate hallucination in vision-language models by using counterfactual analysis to estimate the Natural Direct Effect (NDE) of each modality and their interaction. |
| Outcome: | The proposed framework significantly reduces hallucination while preserving task performance while retaining reliability. |
Improving Word Alignment Using Semi-Supervised Learning (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing word alignment methods rely on labeled data, but augmenting training with pseudo-labeled data improves performance. |
| Approach: | They propose a semi-supervised framework to improve word alignment methods . they use pseudo-labeled data from multilingual encoder models as word aligners . |
| Outcome: | The proposed framework outperforms the current state-of-the-art binary alignment method on word alignment datasets. |
OVFact: Measuring and Improving Open-Vocabulary Factuality for Long Caption Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large vision-language models struggle to generate long and factual captions . traditional measures for hallucination and factuality are not well suited for longer captions. |
| Approach: | They propose a method for measuring caption factuality of long captions that leverages open-vocabulary visual grounding and tool-based verification without relying on human annotations. |
| Outcome: | The proposed method improves agreement with human judgements and captures both caption descriptiveness and factual precision in the same metric. |
Mitigating Intrinsic Named Entity-Related Hallucinations of Abstractive Text Summarization (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Abstractive text summarization (ATS) is important and challenging, but some hallucinations remain a challenge. |
| Approach: | They propose an adaptive margin ranking loss to facilitate two entity-alignment learning methods to tackle named entity-related hallucinations. |
| Outcome: | The proposed method improves the baseline model on automatic evaluation scores. |
Large Language Models with Temporal Reasoning for Longitudinal Clinical Summarization and Prediction (2025.findings-emnlp)
Copied to clipboard
Maya Kruse, Shiyue Hu, Nicholas Derby, Yifu Wu, Samantha Stonbraker, Bingsheng Yao, Dakuo Wang, Elizabeth M. Goldberg, Yanjun Gao
| Challenge: | Recent advances in large language models have shown potential in clinical text summarization, but their ability to handle long patient trajectories with multi-modal data spread across time remains underexplored. |
| Approach: | They evaluate open-source large language models, their Retrieval Augmented Generation variants and chain-of-thought prompting on long-context clinical summarization and prediction. |
| Outcome: | The proposed models can synthesize structured and unstructured EHR data while reasoning over temporal coherence. |
HalluLens: LLM Hallucination Benchmark (2025.acl-long)
Copied to clipboard
Yejin Bang, Ziwei Ji, Alan Schelten, Anthony Hartshorn, Tara Fowler, Cheng Zhang, Nicola Cancedda, Pascale Fung
| Challenge: | Large language models (LLMs) generate responses that deviate from user input or training data, a phenomenon known as "hallucination" . |
| Approach: | They propose a hallucination benchmark HalluLens that includes both extrinsic and intrinsic evaluation tasks to distinguish between extrindic and intrinsic hallucines. |
| Outcome: | The proposed framework disentangles LLM hallucination from "factuality" and distinguishes between extrinsic and intrinsic hallucines to promote consistency and facilitate research. |
Lost in Transcription, Found in Distribution Shift: Demystifying Hallucination in Speech Foundation Models (2025.findings-acl)
Copied to clipboard
| Challenge: | Automatic speech recognition systems have seen remarkable improvements in recent years, but evaluation of performance remains dependent on word and character error rate (WER/CER). |
| Approach: | They investigate how distribution shifts, model size and model architecture influence hallucination error rate (HER) HER is a metric used to quantify hallucinosity in automatic speech recognition systems. |
| Outcome: | The proposed model can be used to measure hallucination errors in high-stakes domains such as healthcare, legal, and aviation. |
Dynamic PMI-Guided Contrastive Decoding Reduces Hallucination in Large Language Models: A Unified Framework of Fine-Grained Input Transformations (2026.findings-acl)
Copied to clipboard
| Challenge: | Despite the remarkable generation capabilities of large language models, the issue of hallucination remains a critical challenge. |
| Approach: | They propose a contrastive decoding framework based on dynamic pointwise mutual information that disentangles spurious dependencies induced by context priors, lexical co-occurrences, and syntactic structures and prioritizes causal logic. |
| Outcome: | The proposed framework significantly improves the model’s factuality and reasoning robustness while maintaining high computational efficiency. |
KaFT: Knowledge-aware Fine-tuning for Boosting LLMs’ Domain-specific Question-Answering Performance (2025.findings-acl)
Copied to clipboard
| Challenge: | Recent literature reveals that supervised fine-tuning (SFT) is suboptimal for domain-specific question-answering tasks. |
| Approach: | They propose a query diversification strategy for robust conflict detection and a knowledge-aware fine-tuning approach to effectively boost LLMs’ performance. |
| Outcome: | The proposed approach improves the model generalization and alleviates the hallucination. |
Adaptively profiling models with task elicitation (2025.emnlp-main)
Copied to clipboard
| Challenge: | Language model evaluations fail to characterize consequential failure modes, forcing experts to inspect outputs and build new benchmarks. |
| Approach: | They propose a method that automatically builds new evaluations to profile model behavior. |
| Outcome: | The proposed method finds that language models fail in hundreds of tasks . it also finds that o3-mini is prone to hallucination when fabrications are repeated . |
Trust Within? Seek Beyond? Knowledge Boundary Aware Policy Optimization for Agentic Search (2026.acl-long)
Copied to clipboard
Tao Feng, Xinke Jiang, Xinyan Hu, Yonggang Zhang, Zhen Tao, Wentao Zhang, Boyang Liu, Wenhao Jiang, Chao Wu
| Challenge: | Existing approaches to augment large language models with external knowledge suffer from a lack of calibration regarding the model’s knowledge boundary. |
| Approach: | They propose a reinforcement learning framework that explicitly aligns retrieval decisions with quantified knowledge states. |
| Outcome: | The proposed framework outperforms strong baselines while exhibiting reduced hallucination rates. |
Controllable and Reliable Knowledge-Intensive Task-Oriented Conversational Agents with Declarative Genie Worksheets (2025.acl-long)
Copied to clipboard
| Challenge: | Existing LLMs suffer from hallucination, following instructions with conditional logic, and integrating knowledge from different sources. |
| Approach: | They propose a programmable framework for creating knowledge-intensive task-oriented conversational agents that handle involved interactions and answer complex queries. |
| Outcome: | The proposed framework outperforms SOTA methods on complex logic dialogue datasets by up to 20.5%. |
ReFLAIR: Enhancing Multimodal Reasoning via Structured Reflection and Reward-Guided Learning (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing training methods for large models do not address the trade-off between reflection and accuracy. |
| Approach: | a unified framework teaches large models to perform structured reflection via an explicit $think re-think answer $ format and hybrid reward learning. |
| Outcome: | The proposed framework improves model performance on mathematical benchmarks and reduces inference cost by nearly 23%. |
Logit Space Constrained Fine-Tuning for Mitigating Hallucinations in LLM-Based Recommender Systems (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing LLM-based recommender systems rely on standard fine-tuning methodologies, often ignoring hallucination issues during the fine-uning process. |
| Approach: | They propose a logit space constraint-based fine-tuning framework to mitigate hallucination in LLM-based recommenders by incorporating Kullback–Leibler divergence into the training objective. |
| Outcome: | Experiments on two recommendation models with distinct LLM backbones and four real-world datasets show that LCFT reduces hallucination and enhances recommendation performance. |
The Agent’s First Day: Benchmarking Learning, Exploration, and Scheduling in the Workplace Scenarios (2026.findings-acl)
Copied to clipboard
Daocheng Fu, Jianbiao Mei, Rong Wu, Xuemeng Yang, Jia Xu, Ding Wang, Pinlong Cai, Yong Liu, Licheng Wen, Botian Shi
| Challenge: | Existing research mainly focuses on performance upper bounds in static environments, overlooking stochastic real-world deployment. |
| Approach: | They propose a dynamic evaluation environment that simulates a "trainee" agent continuously exploring a novel setting. |
| Outcome: | The proposed model evaluates agents in a dynamic evaluation environment that simulates a "trainee" agent continuously exploring a novel setting. |
Anchoring the Cache: Mitigating Contextual Hallucination in KV-Compressed Long-Context Summarization (2026.acl-long)
Copied to clipboard
| Challenge: | Recent studies show that KV cache compression can increase hallucination scores in LLMs . modern LLM models support extremely long sequences, but their impact on model hallucinosity remains underexplored. |
| Approach: | They propose a decoding-phase strategy that selectively removes generated KV pairs from retrieval heads responsible for retrieving critical information from source context. |
| Outcome: | The proposed method reduces hallucination across multiple models and datasets while preserving computational efficiency. |
UNCLE: Benchmarking Uncertainty Expressions in Long-Form Generation (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing work lacks direct and fair evaluation of Large Language Models’ ability to express uncertainty effectively in long-form generation. |
| Approach: | They propose a benchmark to evaluate uncertainty expression in both long- and short-form question answering (QA) they propose prompt-based and training-based methods to improve models’ performance. |
| Outcome: | The proposed method mitigates this issue but a misalignment persists in uncertainty expression between long- and short-form generation. |
CreativeBench: Benchmarking and Enhancing Machine Creativity via Self-Evolving Challenges (2026.findings-acl)
Copied to clipboard
| Challenge: | Increasing saturation of web data limits further scaling of model intelligence. |
| Approach: | They propose a benchmark to evaluate machine creativity in code generation that combines combinatorial and exploratory creativity through reverse engineering and self-play. |
| Outcome: | The proposed benchmark targets combinatorial and exploratory creativity through reverse engineering and self-play. |
HalluAudio: A Comprehensive Benchmark for Hallucination Detection in Large Audio-Language Models (2026.acl-long)
Copied to clipboard
| Challenge: | Existing studies on hallucination focus on text or vision, while few audio-oriented studies are limited in scale, modality coverage, and diagnostic depth. |
| Approach: | They propose a large-scale benchmark for evaluating hallucinations across speech, sound, and music. |
| Outcome: | The proposed model improves hallucination rate, yes/no bias, error-type analysis, and refusal rate. |
Parametric Knowledge is Not All You Need: Toward Honest Large Language Models via Retrieval of Pretraining Data (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language models are highly capable of answering questions, but they are often unaware of their own knowledge boundary, i.e., knowing what they know and what they don’t know. |
| Approach: | They propose a method to evaluate LLM honesty using Pythia with publicly available pretraining data. |
| Outcome: | The proposed method is based on Pythia, a truly open LLM with publicly available pretraining data. |
System-Mediated Attention Imbalances Make Vision-Language Models Say Yes (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing mitigation strategies tend towards an image-centric interpretation of these imbalances, prioritising increased image attention while giving less consideration to the roles of the other modalities. |
| Approach: | They propose a more holistic, system-mediated account which attributes imbalances to functionally redundant system weights that reduce attention to image and textual inputs. |
| Outcome: | The proposed framework offers a useful empirical perspective on the yes-bias, a common form of hallucination in which VLMs indiscriminately respond ‘yes’. |
CausalGaze: Unveiling Hallucinations via Counterfactual Graph Intervention in Large Language Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing classification-based methods capture noise and spurious correlations while overlooking the underlying causal mechanisms. |
| Approach: | They propose a hallucination detection framework based on structural causal models that captures static and passive signals from internal states and employs counterfactual interventions to disentangle causal reasoning paths from incidental noise. |
| Outcome: | Experiments on four datasets and three widely used LLMs show that the proposed framework improves AUROC and interpretability. |
Moneyball with LLMs: Analyzing Tabular Summarization in Sports Narratives (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language model (LLM) approaches to tabular summarization rely on prompt engineering, decomposition pipelines, or entity-level intermediate representations to achieve strong performance. |
| Approach: | They propose a diagnostic benchmark for long-context tabular summarization using decomposition pipelines and entity-level intermediate representations. |
| Outcome: | The proposed benchmark improves accuracy and numerical fidelity, but lacks local arithmetic. |
Paper2Rebuttal: A Multi-Agent Framework for Transparent Author Response Assistance (2026.acl-long)
Copied to clipboard
| Challenge: | Current approaches to writing effective rebuttals are limited by the direct-to-text generation problem . authors must accurately decipher reviewer intent while ensuring every response is firmly anchored in verifiable manuscript details. |
| Approach: | They propose a framework that reframes rebuttal generation as an evidence-centric planning task. |
| Outcome: | The proposed framework outperforms baselines in coverage, faithfulness, and strategic coherence. |
DuET: Dual Execution for Test Output Prediction with Generated Code and Pseudocode (2026.findings-acl)
Copied to clipboard
| Challenge: | Recent studies have shown that test output prediction is difficult to achieve due to code errors. |
| Approach: | They propose a framework that grounds prediction on error-resilient pseudocode and simulates execution via LLM reasoning to overcome limitations of direct execution suffering from code errors. |
| Outcome: | The proposed framework improves Pass@1 on LiveCodeBench, BigCodeBech-Hard, DevEval and HumanEval(+) and improves on pass@1 by 13.6 pp. |
Detecting Hallucinations in SpeechLLMs at Inference Time Using Attention Maps (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for hallucination detection for text-based LLMs do not capture audio-specific signals. |
| Approach: | They propose to capture pathological attention patterns associated with hallucination using four attention-derived metrics to train lightweight logistic regression classifiers. |
| Outcome: | The proposed approach outperforms baselines on in-domain data and generalises to out-of-domain ASR settings. |