Papers with answer accuracy
LogicSolver: Towards Interpretable Math Word Problem Solving with Logical Prompt-enhanced Learning (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Recent advances in MWP solving are uninterpretable due to shallow heuristics . a new approach to solve automatic word problem solvers requires a solver to predict expression tree and corresponding linguistic logic formulas simultaneously. |
| Approach: | They propose to annotate interpretable logical formulas based on algebraic knowledge as the grounded linguistic logic of each solution equation. |
| Outcome: | The proposed approach improves interpretability of a MWP solver by using logical prompts and interpretation generation. |
Discover and Prove: An Open-source Agentic Framework for Hard Mode Automated Theorem Proving in Lean 4 (2026.acl-long)
Copied to clipboard
Chengwu Liu, Yichun Yin, Ye Yuan, Jiaxuan Xie, Botao Li, Siqi Li, Jianhao Shen, Yan Xu, Lifeng Shang, Ming Zhang
| Challenge: | Existing approaches to solving mathematical problems fall into two broad categories: informal methods and formal methods. |
| Approach: | They propose to use LLM natural-language reasoning to discover answers . they introduce Discover And Prove framework that rewrites Hard Mode statements into Easy Mode ones for existing ATP provers. |
| Outcome: | The proposed framework can be used to prove hard mode statements on ATP benchmarks. |
Exploring Generative Process Reward Modeling for Semi-Structured Data: A Case Study of Table Question Answering (2026.eacl-short)
Copied to clipboard
| Challenge: | Recent advances in process reward models (PRMs) have demonstrated remarkable improvements in the reasoning capabilities of large language models (LLMs). |
| Approach: | They evaluate state-of-the-art generative PRMs on table question answering from answer and step perspectives and compare their results to previous studies. |
| Outcome: | The proposed model can aid solution selection but struggle to generalize to out-of-domain data. |
AutoQA: From Databases To QA Semantic Parsers With Only Synthetic Training Data (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to generate semantic parsers that answer questions on databases require large amounts of annotated data. |
| Approach: | They propose a method to generate semantic parsers that answer questions on databases . they use automatic paraphrasing and template-based parsing to find alternative expressions . |
| Outcome: | The proposed method achieves 69.8% answer accuracy on natural questions, 16.4% higher than state-of-the-art models and 5.2% lower than the same model trained with human data. |
Question Decomposition for Retrieval-Augmented Generation (2025.acl-srw)
Copied to clipboard
| Challenge: | Retrieval-augmented generation (RAG) is effective for question answering tasks . multi-hop questions, such as "Which company among NVIDIA, Apple, and Google made the biggest profit in 2023?" challenge RAG because relevant facts are often distributed across multiple documents . |
| Approach: | They propose a pipeline that incorporates question decomposition to ground large language models in verifiable external sources. |
| Outcome: | The proposed approach improves retrieval and answer accuracy over standard RAG . multi-hop questions often require multiple documents to support the model . |
FastV-RAG: Towards Fast and Fine-Grained Video QA with Retrieval-Augmented Generation (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for retrieval-augmented generation are inefficient and often fail to maintain high answer quality. |
| Approach: | They propose an efficient VLM-based RAG framework built on a speculative decoding pipeline and a similarity-based filtering strategy to mitigate errors. |
| Outcome: | The proposed framework reduces inference latency without sacrificing correctness . it achieves comparable or higher accuracy than standard approaches while speeding up inference by approximately 2x . |
Test-Time Strategies for More Efficient and Accurate Agentic RAG (2026.acl-srw)
Copied to clipboard
Abhinav Sharma, Brian Zhang, Deepti Guntur, Zhiyang Zuo, Shreyas Chaudhari, Wenlong Zhao, Franck Dernoncourt, Puneet Mathur, Ryan A. Rossi, Nedim Lipka
| Challenge: | Retrieval-Augmented Generation (RAG) systems face challenges with complex, multi-hop questions. |
| Approach: | They propose to integrate contextualization module and de-duplication module to improve the accuracy of retrieved documents and to reduce the number of turns by 10.5%. |
| Outcome: | The proposed approach achieves a 5.6% increase in EM score and reduces the average number of turns by 10.5% compared to the baseline. |
Granite Guardian: Comprehensive LLM Safeguarding (2025.naacl-industry)
Copied to clipboard
Inkit Padhi, Manish Nagireddy, Giandomenico Cornacchia, Subhajit Chaudhury, Tejaswini Pedapati, Pierre Dognin, Keerthiram Murugesan, Erik Miehling, Martín Santillán Cooper, Kieran Fraser, Giulio Zizzo, Muhammad Zaid Hameed, Mark Purcell, Michael Desmond, Qian Pan, Inge Vejsbjerg, Elizabeth M. Daly, Michael Hind, Werner Geyer, Ambrish Rawat, Kush R. Varshney, Prasanna Sattigeri
| Challenge: | a suite of advanced models is designed to detect and mitigate risks associated with prompts and responses. |
| Approach: | a team of researchers develop a model family to detect and mitigate risks associated with prompts and responses. the model family is based on the Granite 3.0 language models. |
| Outcome: | a new model family is designed to detect and mitigate risks associated with prompts and responses. |
Chatbot Arena Estimate: towards a generalized performance benchmark for LLM capabilities (2025.naacl-industry)
Copied to clipboard
Lucas Spangher, Tianle Li, William F. Arnold, Nick Masiewicki, Xerxes Dotiwalla, Rama Kumar Pasumarthi, Peter Grabowski, Eugene Ie, Daniel Gruhl
| Challenge: | Existing benchmark aggregation methods, such as Elo-based systems, can be resource-intensive, public facing, and time-consuming. |
| Approach: | They propose a framework for aggregating performance across diverse benchmarks that generates a “Goodness” and a ‘Fastness” score. |
| Outcome: | The proposed framework achieves higher Pearson correlation with Chatbot Arena Elo scores than MMLU’s correlation with chatbot Arena scores, validating its reliability for real-world LLM evaluation. |
Pointwise Mutual Information as a Performance Gauge for Retrieval-Augmented Generation (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing methods to improve language models' performance do not exploit this phenomenon . |
| Approach: | They propose to use contextual information to select and construct prompts that improve model performance. |
| Outcome: | The proposed methods show that the mutual information between a context and a question is an effective gauge for language model performance. |
DREAM: Improving Situational QA by First Elaborating the Situation (2022.naacl-main)
Copied to clipboard
| Challenge: | Cognitive science has long promoted the formation of mental models as central to understanding and question-answering. |
| Approach: | They train a new model, DREAM, to answer questions that elaborate the scenes that situated questions are about and then provide those elaborations as additional context to a question-answering (QA) model. |
| Outcome: | The proposed model is able to create better scene elaborations than a representative state-of-the-art, zero-shot model. |
PaperSearchQA: Learning to Search and Reason over Scientific Papers with RLVR (2026.eacl-long)
Copied to clipboard
James Burgess, Jan N. Hansen, Duo Peng, Yuhui Zhang, Alejandro Lozano, Min Woo Sun, Emma Lundberg, Serena Yeung-Levy
| Challenge: | Recent methods supervise only the final answer accuracy using reinforcement learning with verifiable rewards (RLVR). |
| Approach: | They propose to train search agents to search and reason over scientific papers and a factoid QA dataset with 60k biomedical paper abstracts. |
| Outcome: | The proposed model outperforms non-RL retrieval baselines and is scalable and extendable to other scientific domains. |
CondAmbigQA: A Benchmark and Dataset for Conditional Ambiguous Question Answering (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) generate unreliable responses due to their cognitive alignment of context and intent. |
| Approach: | They propose a benchmark to identify possible implicit assumptions in QA questions . they use retrieved Wikipedia fragments to identify interpretations for a given query . |
| Outcome: | The proposed benchmark identifies possible implicit assumptions and improves answer accuracy by 11.75% . retrieved Wikipedia fragments help identify possible interpretations for a given query . |
PAR2-RAG: Planned Active Retrieval and Reasoning for Multi-Hop Question Answering (2026.acl-industry)
Copied to clipboard
| Challenge: | Multi-hop question answering is a practical bottleneck in industry applications . large language models (LLMs) fail frequently when evidence coverage is incomplete or reasoning trajectories drift . |
| Approach: | They propose a training-free two-stage framework that separates coverage from commitment . it performs breadth-first anchoring to build a high-recall evidence frontier . compared with IRCoT, it achieves 23.5% higher answer accuracy . |
| Outcome: | The proposed framework outperforms strong baselines in MHQA benchmarks and achieves 23.5% higher answer accuracy and 10.5% NDCG gains in retrieval quality. |
Towards a Unified Model for Generating Answers and Explanations in Visual Question Answering (2023.findings-eacl)
Copied to clipboard
| Challenge: | Current explanation generation models are trained to select the best answers from Multiple-Choice questions or to classify single-word answers to a predetermined vocabulary. |
| Approach: | They propose a multitask learning approach towards a Unified Model for Answer and Explanation generation (UMAE) UMAE models surpass the prior state-of-the-art answer accuracy on A-OKVQA by 10 15%, show competitive results on OK-VQA and VCR, and demonstrate promising out-of domain performance on VQA-X. |
| Outcome: | The proposed model outperforms the state-of-the-art model on A-OKVQA and VCR and shows promising out-of domain performance on VQA-X. |
MAIN-RAG: Multi-Agent Filtering Retrieval-Augmented Generation (2025.acl-long)
Copied to clipboard
Chia-Yuan Chang, Zhimeng Jiang, Vineeth Rakesh, Menghai Pan, Chin-Chia Michael Yeh, Guanchu Wang, Mingzhi Hu, Zhichao Xu, Yan Zheng, Mahashweta Das, Na Zou
| Challenge: | Existing RAG systems struggle with the quality of retrieval documents, causing performance degradation and reducing performance. |
| Approach: | They propose a training-free RAG framework that leverages multiple LLM agents to collaboratively filter and score retrieved documents. |
| Outcome: | The proposed framework outperforms existing RAG frameworks in QA benchmarks and shows superior answer consistency and answer accuracy over baseline methods. |
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 . |
Confidence-Aware Reasoning: Optimizing Self-Guided Thinking Trajectories in Large Reasoning Models (2025.emnlp-industry)
Copied to clipboard
| Challenge: | Chain-of-thought enables large reasoning models to reason through multi-step problems but often leads to unnecessarily long or redundant reasoning traces, a phenomenon known as overthinking. |
| Approach: | They propose an inference-time framework that selectively prunes low-utility reasoning blocks and halts early when sufficient confidence has been achieved. |
| Outcome: | The proposed framework improves answer accuracy by up to +13.3% while reducing average reasoning length by 40%–50%. |
MRAG: A Modular Retrieval Framework for Time-Sensitive Question Answering (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for answering time-sensitive questions lack temporal reasoning . existing methods struggle with these time-intensive questions, authors say . |
| Approach: | They propose a temporal-based question-answering framework that integrates temporal perturbations and gold evidence labels into a question processing framework. |
| Outcome: | The proposed framework outperforms baseline retrieval methods in retrieval performance. |
Direct Evaluation of Chain-of-Thought in Multi-hop Reasoning with Knowledge Graphs (2024.findings-acl)
Copied to clipboard
| Challenge: | Prior research on evaluating large language models focused on answer accuracy, neglecting the correctness of the generated CoT. |
| Approach: | They propose a discriminative and generative CoT evaluation paradigm to assess LLMs’ knowledge of reasoning and the accuracy of the generated CoT. |
| Outcome: | The proposed evaluation paradigm assesses LLMs’ knowledge of reasoning and the accuracy of the generated CoT. |
DRIVINGVQA: A Dataset for Interleaved Visual Chain-of-Thought in Real-World Driving Scenarios (2026.findings-eacl)
Copied to clipboard
| Challenge: | Chain-of-thought (CoT) prompting is a prompting strategy that improves reasoning in large language models, but its effectiveness in vision-language models remains limited due to over-reliance on textual cues and memorized knowledge. |
| Approach: | They propose a visual question-answering dataset derived from driving theory exams that incorporates textual explanations with visual tokens extracted from entities relevant to the reasoning process. |
| Outcome: | The proposed approach outperforms chain-of-thought prompting in large language models and vision-language models in real-world scenarios. |
SafeConf: A Confidence-Calibrated Safety Self-Evaluation Method for Large Language Models (2025.findings-emnlp)
Copied to clipboard
Bo Zhang, Cong Gao, Linkang Yang, Bingxu Han, Minghao Hu, Zhunchen Luo, Guotong Geng, Xiaoying Bai, Jun Zhang, Wen Yao, Zhong Wang
| Challenge: | Large language models (LLMs) have many advantages but they also pose significant safety risks. |
| Approach: | They propose a method to enhance the safety self-evaluation capability of LLMs . they perform semantic mutations on the original safety evaluation questions . |
| Outcome: | The proposed method improves safety self-evaluation accuracy by 5.86% and 7.79% over baseline methods on Chinese and English datasets. |
Topology-of-Question-Decomposition: Enhancing Large Language Models with Information Retrieval for Knowledge-Intensive Tasks (2025.coling-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) are constrained to chaining immediate reasoning steps and relying solely on parametric knowledge. |
| Approach: | They propose a framework that activates retrieval only when necessary to improve answer accuracy. |
| Outcome: | Experiments show that the proposed framework improves performance in knowledge-intensive tasks. |
Efficient Table Retrieval and Understanding with Multimodal Large Language Models (2026.findings-eacl)
Copied to clipboard
| Challenge: | Tabular data is often captured in image form across a wide range of real-world scenarios. |
| Approach: | They propose a framework that enables MLLMs to answer queries over large tables. |
| Outcome: | The proposed framework outperforms existing methods by 7.0% in retrieval recall and 6.1% in answer accuracy on a newly constructed dataset with 48,504 unique tables. |
Path Spuriousness-aware Reinforcement Learning for Multi-Hop Knowledge Graph Reasoning (2023.eacl-main)
Copied to clipboard
| Challenge: | Multi-hop reasoning is a common approach for query answering, but can be biased to spurious paths which coincidentally lead to the correct answer with poor explanation. |
| Approach: | They propose a method that quantitatively estimates to what extent a path is spurious by a metric called Path Spuriousness (PS) they propose KG reasoning, which infers new facts along existing paths in KGs. |
| Outcome: | The proposed model significantly improves the agent’s ability to prevent spurious paths while keeping comparable to state-of-the-art performance. |
Database reasoning over text (2021.acl-long)
Copied to clipboard
| Challenge: | Existing models cannot handle database queries such as “List/Count all female athletes who were born in 20th century”. |
| Approach: | They propose a modular architecture to answer database-style queries over multiple spans from text and aggregate them at scale. |
| Outcome: | The proposed architecture scales to databases containing thousands of facts whereas current models are limited by how many facts can be encoded. |
Mitigating Selection Bias with Node Pruning and Auxiliary Options (2025.acl-long)
Copied to clipboard
| Challenge: | Large language models exhibit systematic preferences for answer choices when answering multiple-choice questions. |
| Approach: | They propose two methods to identify and remove internal sources of selection bias . they propose Choice Kullback-Leibler Divergence (CKLD) to capture distributional imbalances in model predictions. |
| Outcome: | The proposed methods improve answer accuracy while reducing selection bias. |
A Comprehensive Evaluation of Multilingual Chain-of-Thought Reasoning: Performance, Consistency, and Faithfulness Across Languages (2026.findings-eacl)
Copied to clipboard
| Challenge: | Recent work has examined final-answer accuracy in multilingual settings, but the behavior of thinking traces, i.e., the intermediate steps that lead to the final answer, remains underexplored. |
| Approach: | They propose to measure language compliance, answer accuracy, and answer consistency when LRMs are explicitly instructed or prompt-hacked to think in a target language. |
| Outcome: | The proposed model improves in English and other high-resource languages while relying on traces to varying degrees. |
Efficient Self-Evaluation for Diffusion Language Models via Sequence Regeneration (2026.acl-long)
Copied to clipboard
| Challenge: | Non-sequential and bidirectional nature of diffusion large language models makes direct likelihood-based self-evaluation challenging. |
| Approach: | They propose a self-evaluation confidence quantification method for diffusion large language models that quantifies confidence by computing the probability of regenerating tokens in the entire generated sequence, given the full context. |
| Outcome: | The proposed method is correlated with semantic coherence and answer accuracy. |
Neuro-Symbolic Integration Brings Causal and Reliable Reasoning Proofs (2025.findings-naacl)
Copied to clipboard
| Challenge: | a new framework for complex reasoning with LLMs is developed to improve reasoning proof accuracy and interpretability. |
| Approach: | They propose to use LLMs to generate search logs that can be interpreted into human-readable reasoning proofs. |
| Outcome: | The proposed framework improves reasoning accuracy but lacks interpretability due to black-box nature of the solvers. |
Navigating Large-Scale Document Collections: MuDABench for Multi-Document Analytical QA (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing multi-document QA benchmarks require information from only a few documents with limited cross-document reasoning. |
| Approach: | They propose a benchmark for multi-document analytical QA that extracts and synthesizes information across multiple documents to perform quantitative analysis. |
| Outcome: | The proposed approach improves both process and outcome metrics but still has bottlenecks compared to human experts. |
Fine-tuned LLMs Know More, Hallucinate Less with Few-Shot Sequence-to-Sequence Semantic Parsing over Wikidata (2023.emnlp-main)
Copied to clipboard
| Challenge: | Large language models can answer many questions correctly, but can also hallucinate and give wrong answers. |
| Approach: | They propose a question-answering benchmark for Wikidata that uses SPARQL to ground large language models. |
| Outcome: | The proposed method outperforms the state-of-the-art for QALD-7 by 3.6% in F1 score. |
Over-Searching in Search-Augmented Large Language Models (2026.eacl-long)
Copied to clipboard
| Challenge: | Search-augmented large language models (LLMs) excel at knowledge-intensive tasks by integrating external retrieval. |
| Approach: | They conduct a systematic evaluation of over-searching across multiple dimensions including query types, model categories, retrieval conditions, and multi-turn conversations. |
| Outcome: | The proposed model improves answer accuracy on answerable queries but harms abstention on unanswerable ones . |
Generating Self-Contained and Summary-Centric Question Answer Pairs via Differentiable Reward Imitation Learning (2021.emnlp-main)
Copied to clipboard
| Challenge: | a proposed model for question-answer pairs with self-contained, summary-centric questions and length-constrained, article-summarizing answers is based on suggested question generation in conversational news recommendation systems. |
| Approach: | They propose a model for generating question-answer pairs with self-contained, summary-centric questions and length-constrained, article-summarizing answers. |
| Outcome: | The proposed model captures the central gists of the articles and achieves high answer accuracy. |
ALW: Adaptive Layer-Wise contrastive decoding enhancing reasoning ability in Large Language Models (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing research has demonstrated that contrast decoding of two different models can improve text quality in open-ended text generation but with limited gains on reasoning tasks. |
| Approach: | They propose a framework that dynamically disentangles noise in shallow layers from critical signals in deep layers to enhance reasoning ability. |
| Outcome: | The proposed framework improves answer accuracy while maintaining inference efficiency. |
Refiner: Restructure Retrieved Content Efficiently to Advance Question-Answering Capabilities (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are limited by their parametric knowledge, leading to hallucinations in knowledge-extensive tasks. |
| Approach: | They propose an end-to-end extract-and-restructure paradigm that leverages a single decoder-only LLM to adaptively extract query-relevant contents verbatim along with the necessary context. |
| Outcome: | Experiments show that a trained Refiner outperforms state-of-the-art RAG and compressing approaches in multiple tasks. |
Enhancing Semantics in Multimodal Chain of Thought via Soft Negative Sampling (2024.lrec-main)
Copied to clipboard
| Challenge: | Chain of thought (CoT) is used for complex reasoning problems, but hallucinations are a problem in multimodal CoT. |
| Approach: | They propose a method to generate soft negative samples with different semantics to mitigate hallucinations in multimodal CoT. |
| Outcome: | The proposed method mitigates hallucinations in multimodal CoT by using soft negative sampling. |
Beyond "I Don’t Know": Evaluating LLM Self-Awareness in Discriminating Data and Model Uncertainty (2026.acl-long)
Copied to clipboard
| Challenge: | Prior studies treat refusal as a generic "I don't know" lack of distinction limits downstream action decisions like requesting clarification or invoking external tools. |
| Approach: | They propose a benchmark to evaluate explicit uncertainty attribution in large language models. |
| Outcome: | The proposed method improves uncertainty attribution while preserving answer accuracy. |
ChartGaze: Enhancing Chart Understanding in LVLMs with Eye-Tracking Guided Attention Refinement (2025.emnlp-main)
Copied to clipboard
Ali Salamatian, Amirhossein Abaskohi, Wan-Cyuan Fan, Mir Rayat Imtiaz Hossain, Leonid Sigal, Giuseppe Carenini
| Challenge: | Chart question answering (CQA) is a key research challenge for large vision-language models . recent efforts focus on leveraging LVLMs directly on chart images . |
| Approach: | They propose a gaze-guided attention refinement that aligns image-text attention with human fixations to improve chart reasoning quality and interpretability. |
| Outcome: | The proposed approach improves answer accuracy and attention alignment yielding gains of up to 2.56 percentage points across multiple models. |
When LLMs Read Tables Carelessly: Measuring and Reducing Data Referencing Errors (2026.acl-long)
Copied to clipboard
Yuqing Yang, Qi Zhu, Zhen Han, Boran Han, Zhengyuan Shen, Shuai Wang, Vassilis N. Ioannidis, Huzefa Rangwala
| Challenge: | Large language models (LLMs) perform well on table tasks, but they still make data referencing errors (DREs) prior studies have only offered limited, small-scale analyses. |
| Approach: | They propose inference-time strategies and lightweight critics to mitigate data referencing errors. |
| Outcome: | The proposed model achieves an average F1 score of 78.2% in detecting both in-distribution and out-of-difference DREs and assists inference for larger models. |
SceneAlign: Aligning Multimodal Reasoning to Scene Graphs in Complex Visual Scenes (2026.acl-long)
Copied to clipboard
Chuhan Wang, Xintong Li, Jennifer Yuntong Zhang, Junda Wu, Chengkai Huang, Lina Yao, Julian McAuley, Jingbo Shang
| Challenge: | Existing preference-based approaches fail to address this challenge by exploiting language priors to bypass visual grounding. |
| Approach: | They propose a framework that leverages scene graphs as structured visual information to perform controllable structural interventions. |
| Outcome: | The proposed framework improves answer accuracy and reasoning faithfulness across seven visual reasoning benchmarks. |
Towards Better Understanding of Program-of-Thought Reasoning in Cross-Lingual and Multilingual Environments (2025.findings-acl)
Copied to clipboard
Patomporn Payoungkhamdee, Pume Tuchinda, Jinheon Baek, Samuel Cahyawijaya, Can Udomcharoenchaikit, Potsawee Manakul, Peerat Limkonchotiwat, Ekapol Chuangsuwanich, Sarana Nutanong
| Challenge: | Multi-step reasoning is essential for large language models, yet multilingual performance remains challenging. |
| Approach: | They propose a framework to evaluate Program-of-Thought (PoT) prompting by separating multilingual reasoning from code execution to examine impact of fine-tuning on question-reasoning alignment and reasoning quality. |
| Outcome: | The proposed framework outperforms CoT fine-tuned models in multilingual settings and shows strong correlation between reasoning quality and answer accuracy. |
THOR: A Theta-Gamma Hierarchical Oscillatory Reasoning Framework for Multi-hop QA (2026.acl-long)
Copied to clipboard
| Challenge: | Current large language models (LLMs) have shown a powerful ability of reasoning in understanding question requirement, retrieving the supporting fact and generating a precise answer. |
| Approach: | They propose a brain-inspired Theta-Gamma hierarchical oscillatory reasoning framework which decouples attention between global planning and local retrieval. |
| Outcome: | Extensive comparative experiments and specific validation experiments on multi-hop QA benchmarks show that THOR improves answer accuracy and robustness while mitigating limitations. |
Think Natively: Unlocking Multilingual Reasoning with Consistency-Enhanced Reinforcement Learning (2026.acl-long)
Copied to clipboard
Xue Zhang, Yunlong Liang, Fandong Meng, Songming Zhang, Kaiyu Huang, Yufeng Chen, Xu Jinan, Jie Zhou
| Challenge: | Current Large Reasoning Models exhibit two critical limitations when processing non-English languages: (1) They struggle to maintain input-output language consistency; (2) They generally perform poorly with wrong reasoning paths and lower answer accuracy compared to English. |
| Approach: | They propose a language-consistency reward and a cross-lingual thinking alignment reward to improve the model's interpretability and accuracy. |
| Outcome: | The proposed model achieves nearly 100% language consistency and superior performance on two multilingual benchmarks (MMATH and PolyMath). |
Pardon? Evaluating Conversational Repair in Large Audio-Language Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing evaluations of large audio-language models focus on answer accuracy and robustness to acoustic perturbations, but they assume that inputs remain semantically answerable. |
| Approach: | They propose a repair-aware evaluation setting that explicitly distinguishes between answerable and unanswerable audio inputs. |
| Outcome: | The proposed evaluation setting distinguishes between answerable and unanswerable audio inputs. |
Embedding Domain Knowledge for Large Language Models via Reinforcement Learning from Augmented Generation (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to embed knowledge into large language models have some limitations . static nature of training data and lack of knowledge in domains create knowledge gaps . |
| Approach: | They propose a method that iteratively cycles between sampling generations and optimizing the model through calculated rewards. |
| Outcome: | The proposed method outperforms baseline approaches on medical, legal, astronomy, and current events datasets. |
GRACE: Discriminator-Guided Chain-of-Thought Reasoning (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing language models (LMs) can assign a high likelihood to incorrect steps . Existing models (LLMs), however, struggle with complex multi-step reasoning. |
| Approach: | They propose a stepwise decoding approach that steers the decoding process towards producing correct reasoning steps. |
| Outcome: | The proposed approach outperforms existing methods on math and symbolic reasoning tasks. |
When Models Reason in Your Language: Controlling Thinking Language Comes at the Cost of Accuracy (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Recent Large Reasoning Models (LRMs) with thinking traces have shown strong performance on English reasoning tasks. |
| Approach: | They evaluate two leading LRMs with thinking traces on established benchmark XReasoning and propose directions for future research. |
| Outcome: | The proposed models often revert to English or produce fragmented reasoning in other languages, revealing a substantial gap in the capability of thinking in non-English languages. |
CharTide: Data-Centric Chart-to-Code Generation via Tri-Perspective Tuning and Inquiry-Driven Evolution (2026.acl-long)
Copied to clipboard
Xiangxi Zheng, Kuang He, Jiayi Hu, Ping Yu, Rui Yan, Yuan Yao, Peng Hou, Anxiang Zeng, Alex Jinpeng Wang
| Challenge: | Existing approaches to chart-to-code generation are constrained by data-centric limitations . authors present a new framework that redesigns both training and alignment data . |
| Approach: | They propose a data-centric framework that redesigns both training and alignment data for chart-to-code generation. |
| Outcome: | The proposed framework outperforms open-source baselines and is competitive with GPT-5. |
Reason Only When Needed: Efficient Generative Reward Modeling via Model-Internal Uncertainty (2026.findings-acl)
Copied to clipboard
Chao Xue, Yao Wang, Mengqiao Liu, Di Liang, Xingsheng Han, Peiyang Liu, Xianjie Wu, Chenyao Lu, Lei Jiang, Yu Lu, Haibo Shi, Shuang Liang, Minlong Peng, Flora D. Salim
| Challenge: | Existing approaches to generating reward models rely on voting-based mechanisms to evaluate CoT outputs. |
| Approach: | They propose an efficient generative reward modeling framework grounded in model-internal uncertainty. |
| Outcome: | The proposed framework reduces inference cost while improving answer accuracy. |
Stable-RAG: Mitigating Retrieval-Permutation-Induced Hallucinations in Retrieval-Augmented Generation (2026.acl-long)
Copied to clipboard
| Challenge: | Existing RAG methods focus on enhancing LLM robustness to low-quality retrieval, but neither address permutation sensitivity. |
| Approach: | They propose a method that exploits permutation sensitivity to mitigate hallucinations in Large Language Models. |
| Outcome: | The proposed model improves answer accuracy, reasoning consistency, and generalization across datasets, retrievers, and input lengths compared with strong baselines. |
Dynamic Infilling Anchors for Format-Constrained Generation in Diffusion Large Language Models (2026.acl-long)
Copied to clipboard
| Challenge: | Diffusion large language models (dLLMs) offer bidirectional attention and parallel generation . fixed anchors can enforce constraints, but they often impose rigid spans, leading to truncated reasoning . |
| Approach: | They propose a method that dynamically estimates end-anchor positions to adjust generation length before iterative infilling. |
| Outcome: | The proposed method improves format compliance and answer accuracy on GSM8K and MATH. |
BloomEval: A Bloom’s Cognitive Taxonomy-Based Benchmark for Evaluating LRMs via Cognitive Hierarchy Trace (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing benchmarks for Large Reasoning Models rely on answer correctness, but fail to assess the structural coherence and cognitive soundness of the reasoning process itself. |
| Approach: | They propose a framework that maps a model's reasoning trajectory onto hierarchical cognitive levels and an annotation pipeline to ensure a scalable yet reliable annotation pipeline. |
| Outcome: | The proposed framework detects hierarchy jumps, breaks, and overthinking errors and enables scalable yet reliable annotation. |
Towards semantic reliable clinical QA: Query pipeline optimization for cancer patient question answering systems (2026.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are promising for medical Question-Answering but suffer from hallucinations that jeopardize patient safety. |
| Approach: | They propose a three-level controllable metadata-aware framework optimized for Cancer Patient QA (CPQA) they propose combining semantic retrieval with real-time Boolean search to overcome metadata blindness. |
| Outcome: | The proposed framework improves the answer accuracy of Claude-3-haiku by 5.24% over chain-of-thought prompting and about 3% over a naive RAG setup. |
DualRAG: A Dual-Process Approach to Integrate Reasoning and Retrieval for Multi-Hop Question Answering (2025.acl-long)
Copied to clipboard
Rong Cheng, Jinyi Liu, Yan Zheng, Fei Ni, Jiazhen Du, Hangyu Mao, Fuzheng Zhang, Bo Wang, Jianye Hao
| Challenge: | Existing approaches to multi-hop question answering struggle to identify and organize dynamic knowledge . et al., 2023; Liu e.t. al. 2023) suggest a dual-process framework for multi-step reasoning . |
| Approach: | They propose a synergistic dual-process framework that integrates reasoning and retrieval. |
| Outcome: | The proposed framework improves answer accuracy and coherence even in smaller-scale models. |
Progressive Re-ranking for Multimodal Retrieval-Augmented Generation via Curriculum Learning (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing approaches to improve retrieval performance of large language models are limited by static knowledge. |
| Approach: | They propose a multimodal re-ranking framework that combines curriculum learning with fine-grained reranking and multimodal section reassessment to improve CLIP-based visual coarse-grain retrieval. |
| Outcome: | The proposed framework achieves state-of-the-art answer accuracy and competitive retrieval performance on InfoSeek and Enc-VQA. |