Papers by Qin Zhao
Invoke Interfaces Only When Needed: Adaptive Invocation for Large Language Models in Question Answering (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a new metric is developed to pinpoint the moment of invocation when hallucinations arise in small LMs. |
| Approach: | They propose a metric that measures hallucinations during the generation process of small LMs. |
| Outcome: | The proposed metric outperforms baselines in hallucination detection across multiple QA datasets. |
Make Your Decision Convincing! A Unified Two-Stage Framework: Self-Attribution and Decision-Making (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing frameworks for explaining black-box model behavior are unreliable . large-scale pre-trained models often rely on superficial clues for predictions . |
| Approach: | They propose a unified two-stage framework that uses subsequences from the input text as a rationale to generate model decision. |
| Outcome: | The proposed framework achieves competitive results on five reasoning datasets and in semi-supervised scenarios. |
The Design and Construction of a Chinese Sarcasm Dataset (2020.lrec-1)
Copied to clipboard
| Challenge: | Existing sarcasm datasets are limited to English and Chinese . sarcasm is a multi-layered semi-conscious language phenomenon . |
| Approach: | They propose to build a high-quality Chinese sarcasm dataset using user comments . they use manual annotated sarkastic texts and non-sarcastic texts to train sarcasm classifier . |
| Outcome: | The proposed dataset contains 2,486 manual annotated sarcastic texts and 89,296 non-sarcatic texts. |
LLMs Assist NLP Researchers: Critique Paper (Meta-)Reviewing (2024.emnlp-main)
Copied to clipboard
Jiangshu Du, Yibo Wang, Wenting Zhao, Zhongfen Deng, Shuaiqi Liu, Renze Lou, Henry Zou, Pranav Narayanan Venkit, Nan Zhang, Mukund Srinath, Haoran Zhang, Vipul Gupta, Yinghui Li, Tao Li, Fei Wang, Qin Liu, Tianlin Liu, Pengzhi Gao, Congying Xia, Chen Xing, Cheng Jiayang, Zhaowei Wang, Ying Su, Raj Shah, Ruohao Guo, Jing Gu, Haoran Li, Kangda Wei, Zihao Wang, Lu Cheng, Surangika Ranathunga, Meng Fang, Jie Fu, Fei Liu, Ruihong Huang, Eduardo Blanco, Yixin Cao, Rui Zhang, Philip Yu, Wenpeng Yin
| Challenge: | a comparative analysis of paper (meta-)reviews by large language models (LLMs) aims to identify and distinguish LLMs from human activities . |
| Approach: | They present a comparative analysis to identify and distinguish LLM activities from human activities. |
| Outcome: | The proposed analysis aims to improve recognition of instances when someone implicitly uses LLMs for reviewing activities. |
AS-ES Learning: Towards efficient CoT learning in small models (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to induce Chain-of-Thought (CoT) in LLMs are limited and do not consider the importance of efficiently utilizing existing CoT data. |
| Approach: | They propose a new training paradigm which exploits the inherent information in CoT for iterative generation. |
| Outcome: | The proposed training paradigm surpasses direct seq2seq training on CoT-extensive tasks without data augmentation or altering the model itself. |
An Efficient Task-Oriented Dialogue Policy: Evolutionary Reinforcement Learning Injected by Elite Individuals (2025.acl-long)
Copied to clipboard
| Challenge: | Evolutionary Algorithms (EAs) have been proven to effectively explore the solution space of neural networks by maintaining population diversity. |
| Approach: | They propose an elite individual injection mechanism to enhance EA’s search efficiency by adaptively introducing best-performing individuals into the population. |
| Outcome: | Experiments on four datasets show that the proposed approach significantly improves the balance between exploration and exploitation, boosting performance. |
When Correct Beliefs Collapse: Epistemic Resilience of LLMs under Clinical Pressure (2026.acl-long)
Copied to clipboard
| Challenge: | Existing models exhibit severe multi-turn sycophancy in clinical dialogue . high initial diagnostic capability does not imply high belief stability . |
| Approach: | They propose a stress test framework that evaluates belief stability under escalating pressure. |
| Outcome: | The proposed stress test framework reduces the risk of multi-turn sycophancy in clinical dialogue . it eliminates belief change and improves robustness in training time . |
Revealing Personality Traits: A New Benchmark Dataset for Explainable Personality Recognition on Dialogues (2024.emnlp-main)
Copied to clipboard
| Challenge: | Current research treats personality recognition as a classification task, failing to reveal the supporting evidence for the recognized personality. |
| Approach: | They propose a task that aims to reveal the reasoning process as supporting evidence of the personality trait. |
| Outcome: | The proposed task reveals the reasoning process as supporting evidence of the personality trait. |
SimulSpeech: End-to-End Simultaneous Speech to Text Translation (2020.acl-main)
Copied to clipboard
| Challenge: | SimulSpeech is an end-to-end simultaneous speech to text translation system . conventional approaches to simultaneous speech translation divide the translation process into two stages . |
| Approach: | They develop an end-to-end simultaneous speech to text translation system which translates speech in source language to text in target language concurrently. |
| Outcome: | The proposed system achieves reasonable BLEU scores and lower delay compared to full-sentence translation model. |
CrowdAgent: Multi-Agent Managed Multi-Source Annotation System (2025.emnlp-demos)
Copied to clipboard
Maosheng Qin, Renyu Zhu, Mingxuan Xia, null Chenchenkai, Zhen Zhu, Minmin Lin, Junbo Zhao, Lu Xu, Changjie Fan, Runze Wu, Haobo Wang
| Challenge: | Recent approaches to annotate data focus on labeling, but lack holistic process control . a novel system that integrates task assignment, data annotation, and quality/cost management is needed . |
| Approach: | They propose a multi-agent system that integrates task assignment, data annotation, and quality/cost management. |
| Outcome: | The proposed system automates human management by using a collaborative multi-agent system. |
Revisiting Over-Smoothness in Text to Speech (2022.acl-long)
Copied to clipboard
| Challenge: | Non-autoregressive text to speech models ignore correlation in time and frequency domains, causing blurry results. |
| Approach: | They revisit the problem of over-smoothness in non-autoregressive text to speech models . they use methods that reduce complexity of data distributions and improve modeling methods . |
| Outcome: | The proposed models achieve better voice quality and faster inference speed than autoregressive models. |
Verify-and-Edit: A Knowledge-Enhanced Chain-of-Thought Framework (2023.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) have a number of shortcomings, including lack of factual correctness. |
| Approach: | They propose a framework to increase prediction factuality by post-editing reasoning chains . they propose to use large language models to generate interpretable reasoning chains. |
| Outcome: | The proposed framework leads to accuracy improvements in open-domain question-answering tasks. |
Beta Distribution Guided Aspect-aware Graph for Aspect Category Sentiment Analysis with Affective Knowledge (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for aspect category sentiment analysis do not necessarily occur in a sentence. |
| Approach: | They propose a Beta Distribution-guided aspect-aware graph construction based on external knowledge . they use aspect-related words as the pivots to derive aspect-relevant weights . |
| Outcome: | The proposed approach outperforms the state-of-the-art methods on 6 benchmark datasets. |
RedCoder: Automated Multi-Turn Red Teaming for Code LLMs (2026.acl-long)
Copied to clipboard
Wenjie Jacky Mo, Qin Liu, Xiaofei Wen, Dongwon Jung, Hadi Askari, Wenxuan Zhou, Zhe Zhao, Muhao Chen
| Challenge: | Existing red-teaming approaches for code generation rely on extensive human effort and are prone to generating malicious code under adversarial environments. |
| Approach: | They propose a red-teaming agent that engages victim models in multi-turn conversations to elicit vulnerable code. |
| Outcome: | Experiments show that RedCoder outperforms red-teaming methods in inducing vulnerabilities in code generation. |
Beware of Your Po! Measuring and Mitigating AI Safety Risks in Role-Play Fine-Tuning of LLMs (2025.acl-long)
Copied to clipboard
Weixiang Zhao, Yulin Hu, Yang Deng, Jiahe Guo, Xingyu Sui, Xinyang Han, An Zhang, Yanyan Zhao, Bing Qin, Tat-Seng Chua, Ting Liu
| Challenge: | Existing role-play fine-tuning techniques improve role adaptability but may degrade safety performance, especially for villainous characters. |
| Approach: | They propose safety-aware Role-Play Fine-Tuning (SaRFT) to balance role-playing capabilities and safety. |
| Outcome: | The proposed method outperforms state-of-the-art baselines under both LoRA and full-parameter fine-tuning settings. |
Retrieving Multimodal Information for Augmented Generation: A Survey (2023.findings-emnlp)
Copied to clipboard
Ruochen Zhao, Hailin Chen, Weishi Wang, Fangkai Jiao, Xuan Long Do, Chengwei Qin, Bosheng Ding, Xiaobao Guo, Minzhi Li, Xingxuan Li, Shafiq Joty
| Challenge: | Large Language Models (LLMs) are increasingly using multimodality to augment their generation ability, but there is no unified perception of at which stage and how to incorporate different modalities. |
| Approach: | They propose to use multimodality to augment Large Language Models (LLMs) this will provide scholars with a deeper understanding of the methods' applications and encourage them to adapt existing techniques to the fast-growing field of LLMs. |
| Outcome: | The proposed methods improve factuality, reasoning, interpretability, and robustness of the generated content. |
UniCoRN: Unified Cognitive Signal ReconstructioN bridging cognitive signals and human language (2023.acl-long)
Copied to clipboard
| Challenge: | Existing studies focus on decoding word-level fMRI volumes from a restricted vocabulary. |
| Approach: | They propose an open-vocabulary task to bridge fMRI time series and human language . they use a pre-trained language model to construct a robust encoder for cognitive signals . |
| Outcome: | The proposed task bridges fMRI time series and human language with a baseline model. |
Collaborative Chain-of-Agents for Parametric-Retrieved Knowledge Synergy (2026.acl-long)
Copied to clipboard
| Challenge: | Existing RAG methods focus on external retrieval, while ignoring the rich content of the model. |
| Approach: | They propose a framework that enhances explicit synergy over parametric and retrieved knowledge by integrating external retrieval components into the input context of the LLMs. |
| Outcome: | The proposed framework enhances explicit synergy over parametric and retrieved knowledge. |
Retrieve, Discriminate and Rewrite: A Simple and Effective Framework for Obtaining Affective Response in Retrieval-Based Chatbots (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Existing work on retrieval-based chatbots has low-quality affect response . Existing frameworks for obtaining affective response are based on Retrieve-and-Rerank . |
| Approach: | They propose a retrieval-based framework which provides affective response for retrieval chatbots by using a new discriminate-and-rewrite mechanism. |
| Outcome: | The proposed framework outperforms existing baselines and can guarantee the quality of the response and satisfy the affect label. |
MMGCN: Multimodal Fusion via Deep Graph Convolution Network for Emotion Recognition in Conversation (2021.acl-long)
Copied to clipboard
| Challenge: | Emotion recognition in conversation is a crucial component in affective dialogue systems, which helps the system understand users’ emotions and generate empathetic responses. |
| Approach: | They propose a multimodal fused graph convolutional network model which leverages multimodal dependencies and speaker information to model inter-speaker and intra-speech dependency. |
| Outcome: | The proposed model outperforms other SOTA methods on two public benchmark datasets, IEMOCAP and MELD. |
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. |
Texar: A Modularized, Versatile, and Extensible Toolkit for Text Generation (P19-3)
Copied to clipboard
Zhiting Hu, Haoran Shi, Bowen Tan, Wentao Wang, Zichao Yang, Tiancheng Zhao, Junxian He, Lianhui Qin, Di Wang, Xuezhe Ma, Zhengzhong Liu, Xiaodan Liang, Wanrong Zhu, Devendra Sachan, Eric Xing
| Challenge: | Texar is an open-source text generation toolkit that supports a broad set of text generation tasks. |
| Approach: | They introduce Texar, an open-source text generation toolkit that supports text generation tasks. |
| Outcome: | Texar supports machine translation, summarization, dialog, content manipulation, and more. |
Beyond Frameworks: Unpacking Collaboration Strategies in Multi-Agent Systems (2025.acl-long)
Copied to clipboard
| Challenge: | Existing frameworks prioritize structural architectures and role assignments but neglect granular mechanics of agent collaboration. |
| Approach: | They propose to use centralized governance, instructor-led participation, ordered interaction patterns to optimize task accuracy and computational efficiency. |
| Outcome: | The proposed model improves task accuracy and computational efficiency under two context-dependent scenarios. |
Consolidation or Adaptation? PRISM: Disentangling SFT and RL Data via Gradient Concentration (2026.acl-long)
Copied to clipboard
Yang Zhao, Yangou Ouyang, Xiao Ding, Hepeng Wang, Bibo Cai, Kai Xiong, Jinglong Gao, Zhouhao Sun, Li Du, Bing Qin, Ting Liu
| Challenge: | Existing data arbitration strategies for large language model training rely on surface-level heuristics that fail to diagnose intrinsic learning needs. |
| Approach: | They propose a framework that arbitrates data based on its degree of cognitive conflict with the model's existing knowledge. |
| Outcome: | Extensive experiments on WebShop and ALFWorld show that PRISM outperforms state-of-the-art hybrid methods while reducing computational costs by up to 3.22 . |
Diversifying Content Generation for Commonsense Reasoning with Mixture of Knowledge Graph Experts (2022.findings-acl)
Copied to clipboard
| Challenge: | Recent years have seen a surge of interest in improving the generation quality of commonsense reasoning tasks. |
| Approach: | They propose a method that diversifies the generative reasoning by a mixture of expert strategy on commonsense knowledge graphs to encourage various generation outputs. |
| Outcome: | The proposed method improves diversity while achieving on par performance on two GCR benchmarks, based on both automatic and human evaluations. |
Efficient Sequence Learning with Group Recurrent Networks (N18-1)
Copied to clipboard
| Challenge: | Recurrent neural networks have achieved state-of-the-art results in many artificial intelligence tasks, such as language modeling, neural machine translation and speech recognition. |
| Approach: | They propose an efficient architecture to improve the efficiency of such RNN model training by adopting the group strategy for recurrent layers while exploiting the representation rearrangement strategy between layers as well as time steps. |
| Outcome: | The proposed architecture achieves comparable or better accuracy compared with baselines, with a much smaller number of parameters and at a lower computational cost. |
Beyond Similarity: A Gradient-based Graph Method for Instruction Tuning Data Selection (2025.acl-long)
Copied to clipboard
Yang Zhao, Li Du, Xiao Ding, Yangou Ouyang, Hepeng Wang, Kai Xiong, Jinglong Gao, Zhouhao Sun, Dongliang Xu, Qing Yang, Dongchen Li, Bing Qin, Ting Liu
| Challenge: | Existing methods for selecting training data from general datasets fail to account for the joint distribution of instructions, resulting in inefficient learning and suboptimal knowledge transfer. |
| Approach: | They propose a method that constructs a mixed gradient-based instruction graph to capture the joint distribution and interdependencies among instructions. |
| Outcome: | The proposed method outperforms existing methods on domain adaptation tasks and in complex, data-scarce scenarios. |
An Iterative Emotion Interaction Network for Emotion Recognition in Conversations (2020.coling-main)
Copied to clipboard
| Challenge: | Emotion recognition in conversations (ERC) is a task that aims to recognize the emotion of each utterance in conversations. |
| Approach: | They propose an iterative emotion interaction network which uses iterativly predicted emotion labels instead of gold emotion labels to explicitly model the emotion interaction. |
| Outcome: | The proposed method retains state-of-the-art performance on two datasets and achieves high accuracy. |
ChildMandarin: A Comprehensive Mandarin Speech Dataset for Young Children Aged 3-5 (2025.acl-long)
Copied to clipboard
Jiaming Zhou, Shiyao Wang, Shiwan Zhao, Jiabei He, Haoqin Sun, Hui Wang, Cheng Liu, Aobo Kong, Yujie Guo, Xi Yang, Yequan Wang, Yonghua Lin, Yong Qin
| Challenge: | Automatic speech recognition systems have advanced significantly with models like Whisper, Conformer, and self-supervised frameworks such as Wav2vec 2.0. |
| Approach: | They propose to use Mandarin speech datasets to analyze pronunciation and tone of children aged 3 to 5 and evaluate their models on speaker verification (SV) They find that the datasets are more robust than those used by adult speech recognition systems and are open-source and available for all academic purposes. |
| Outcome: | The proposed dataset includes 41.25 hours of speech with carefully crafted manual transcriptions, collected from 397 speakers across various provinces in China, with balanced gender representation. |
Semantic-Aware Action Space Compression via LLM-DRL Synergy for Efficient Task-oriented Dialogue Policy Exploration (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Pre-trained large language models (LLMs) with world knowledge and semantic understanding are promising for task-oriented dialogue systems. |
| Approach: | a framework that synergizes pre-trained large language models with DRL is proposed . a lightweight action pruning mechanism is employed to eliminate implausible actions . |
| Outcome: | a new framework synergizes pre-trained large language models with DRL to guide decision-making . the proposed framework eliminates semantically implausible or low-potential actions from multi-turn dialogue context . |
SSR: Utilizing Simplified Stance Reasoning Process for Robust Stance Detection (2022.coling-1)
Copied to clipboard
| Challenge: | Existing methods for stance detection are task-agnostic, which fail to utilize task knowledge to better discriminate between genuine and bias features. |
| Approach: | They propose to incorporate stance reasoning process as task knowledge to aid in learning genuine features without using targets. |
| Outcome: | The proposed model achieves better performance than previous task-agnostic debiasing methods on new test sets. |
From Hypothesis to Publication: A Comprehensive Survey of AI-Driven Research Support Systems (2025.findings-emnlp)
Copied to clipboard
Zekun Zhou, Xiaocheng Feng, Lei Huang, Xiachong Feng, Ziyun Song, Ruihan Chen, Liang Zhao, Weitao Ma, Yuxuan Gu, Baoxin Wang, Dayong Wu, Guoping Hu, Ting Liu, Bing Qin
| Challenge: | rapid development of artificial intelligence (AI) technologies has inspired researchers to explore how AI can accelerate and enhance research. |
| Approach: | They organize the relevant studies into three main categories: hypothesis formulation, hypothesis validation, and manuscript publication. |
| Outcome: | The authors summarize the current state of research in three main areas: hypothesis formulation, hypothesis validation, and manuscript publication. |
On Safety Risks in Experience-Driven Self-Evolving Agents (2026.findings-acl)
Copied to clipboard
Weixiang Zhao, Yichen Zhang, Yingshuo Wang, Yang Deng, Yanyan Zhao, Xuda Zhi, Yongbo Huang, Hao He, Wanxiang Che, Bing Qin, Ting Liu
| Challenge: | Experience-driven self-evolution has emerged as a promising paradigm for improving the autonomy of large language model agents, yet its reliance on self-curated experience introduces underexplored safety risks. |
| Approach: | They investigate how experience accumulation and utilization in self-evolving agents affect safety performance across web-based and embodied environments. |
| Outcome: | The findings expose inherent limitations of current self-evolving agents and call for more principled strategies to ensure safe and reliable adaptation. |
DIFFA-2: A Practical Diffusion Large Language Model for General Audio Understanding (2026.findings-acl)
Copied to clipboard
| Challenge: | Autoregressive (AR) large audio language models are expensive in data and computation . prior work shows diffusion-based LALMs can improve audio understanding under matched settings . |
| Approach: | They propose a diffusion-based LALM that upgrades the speech encoder and employs dual semantic and acoustic adapters. |
| Outcome: | a new model improves over existing autoregressive large language models and is competitive to strong AR models . the proposed model can make use of limited training data and improve inference efficiency . a recent study shows that diffusion-based models can improve audio understanding . |
MPO: Multilingual Safety Alignment via Reward Gap Optimization (2025.acl-long)
Copied to clipboard
Weixiang Zhao, Yulin Hu, Yang Deng, Tongtong Wu, Wenxuan Zhang, Jiahe Guo, An Zhang, Yanyan Zhao, Bing Qin, Tat-Seng Chua, Ting Liu
| Challenge: | Existing preference learning methods for safety alignment are monolingual and struggle with noisy multilingual data. |
| Approach: | They propose a multilingual reward gaP optimization approach that leverages the well-aligned safety capabilities of the dominant language to improve safety alignment across multiple languages. |
| Outcome: | Extensive experiments on three LLMs, LLaMA-3.1, Gemma-2 and Qwen2.5, validate MPO’s efficacy in multilingual safety alignment without degrading general multilingual utility. |
Automatic and Reliable Evaluation for Academic Caption-to-Figure Generation with LMMs (2026.acl-long)
Copied to clipboard
| Challenge: | Existing datasets for evaluating text-to-image generation focus mostly on real-life images, which poses challenges for assessing academic figure generation given real scientific captions. |
| Approach: | They propose a dataset that first provides a Holistic Evaluation for Academic caption-to-Figure Generation (HE4AFG) they collect real figure captions from 8 scientific domains and generate 3,900 evaluation samples . |
| Outcome: | The proposed model provides high-quality human ratings in terms of three aspects—scientific aesthetic (SA), topic relevance (TR), and attribute correctness (AC). |
Alleviating Hallucinations from Knowledge Misalignment in Large Language Models via Selective Abstention Learning (2025.acl-long)
Copied to clipboard
Lei Huang, Xiaocheng Feng, Weitao Ma, Yuchun Fan, Xiachong Feng, Yuxuan Gu, Yangfan Ye, Liang Zhao, Weihong Zhong, Baoxin Wang, Dayong Wu, Guoping Hu, Lingpeng Kong, Tong Xiao, Ting Liu, Bing Qin
| Challenge: | Large language models (LLMs) suffer from severe hallucination issues due to the knowledge misalignment between the pre-training stage and the supervised fine-tuning stage. |
| Approach: | They propose a training objective with an abstention mechanism that selectively rejects tokens that misalign with the desired knowledge distribution via a special [REJ] token. |
| Outcome: | The proposed model selectively rejects tokens that misalign with the desired knowledge distribution via a special [REJ] token. |
LongLeader: A Comprehensive Leaderboard for Large Language Models in Long-context Scenarios (2025.naacl-long)
Copied to clipboard
Pei Chen, Hongye Jin, Cheng-Che Lee, Rulin Shao, Jingfeng Yang, Mingyu Zhao, Zhaoyu Zhang, Qin Lu, Kaiwen Men, Ning Xie, Huasheng Li, Bing Yin, Han Li, Lingyun Wang
| Challenge: | LongLeader aims to assess different LLMs' long-context comprehension abilities . long-constext comprehension is a key bottleneck for many use cases . |
| Approach: | They propose a leaderboard to assess different LLMs' long-context comprehension abilities . they offer open-source access to the benchmarks and maintain a dedicated website . |
| Outcome: | The proposed model assesses different LLMs on selected benchmarks and provides open-source access to the benchmarks. |
Automatic Article Commenting: the Task and Dataset (P18-2)
Copied to clipboard
| Challenge: | Existing methods to make comments on articles are based on human-annotated subsets, but they are not suitable for online forums. |
| Approach: | They propose to use a large-scale Chinese corpus with millions of real comments and a human-annotated subset characterizing the comments’ varying quality to generalize a broad set of popular reference-based metrics. |
| Outcome: | The proposed model incorporates human-annotated subset characterizing the comments’ varying quality and shows that it is more accurate than previous models. |
When Personalization Legitimizes Risks: Uncovering Safety Vulnerabilities in Personalized Dialogue Agents (2026.acl-long)
Copied to clipboard
Jiahe Guo, Xiangran Guo, Yulin Hu, Zimo Long, Xingyu Sui, Xuda Zhi, Yongbo Huang, Hao He, Weixiang Zhao, Yanyan Zhao, Bing Qin
| Challenge: | Existing research on personalized LLM agents focuses on the effectiveness of personalized responses. |
| Approach: | They propose a benchmark to quantify intent legitimation in personalized interactions . they propose 'detection-reflection' method that detects intent legititimation from internal representation space . |
| Outcome: | The proposed method reduces safety degradation by using internal representation space. |
Both Matter: Enhancing the Emotional Intelligence of Large Language Models without Compromising the General Intelligence (2024.findings-acl)
Copied to clipboard
| Challenge: | Emotional Intelligence (EI) is a key concept in the field of human intelligence. |
| Approach: | They propose a method to enhance EI of large language models by naive fine-tuning on EI-related tasks. |
| Outcome: | The proposed method improves EI of two LLM-based assistants without compromising GI. |
Sentiment Word Aware Multimodal Refinement for Multimodal Sentiment Analysis with ASR Errors (2022.findings-acl)
Copied to clipboard
| Challenge: | Existing models for multimodal sentiment analysis are limited in their capacity to be deployed in the real world. |
| Approach: | They propose a model that can dynamically refine erroneous sentiment words by leveraging multimodal sentiment clues. |
| Outcome: | The proposed model surpasses the state-of-the-art models on three datasets. |
MLDebugging: Towards Benchmarking Code Debugging Across Multi-Library Scenarios (2025.findings-acl)
Copied to clipboard
JinYang Huang, Xiachong Feng, Qiguang Chen, Hanjie Zhao, Zihui Cheng, Jiesong Bai, Jingxuan Zhou, Min Li, Libo Qin
| Challenge: | MLDebugging is a benchmark designed to assess debugging challenges within multi-library Python code. |
| Approach: | They propose to introduce a benchmark to assess debugging challenges within multi-library Python code using 126 Python libraries. |
| Outcome: | The proposed benchmark covers 126 Python libraries and a wide range of multi-library code issues. |
MAESTRO: Meta-learning Adaptive Estimation of Scalarization Trade-offs for Reward Optimization (2026.acl-long)
Copied to clipboard
Yang Zhao, Hepeng Wang, Xiao Ding, Yangou Ouyang, Bibo Cai, Kai Xiong, Jinglong Gao, Zhouhao Sun, Li Du, Bing Qin, Ting Liu
| Challenge: | Group-Relative Policy Optimization (GRPO) has emerged as an efficient paradigm for aligning Large Language Models (LLMs), but its efficacy is confined to domains with verifiable ground truths. |
| Approach: | They propose a meta-cognitive orchestration layer that treats reward scalarization as a dynamic latent policy, leveraging the model’s terminal hidden states as 'a semantic bottleneck' . Across seven benchmarks, MAESTRO consistently outperforms single-reward and static multi-objective baselines while preserving the efficiency advantages of GRPO. |
| Outcome: | The proposed model outperforms single-reward and static multi-objective baselines while preserving efficiency advantages. |
PsychEval: A Multi-Session and Multi-Therapy Benchmark for High-Realism AI Psychological Counselor (2026.findings-acl)
Copied to clipboard
Qianjun Pan, Junyi Wang, Jie Zhou, Yutao Yang, Junsong Li, Kaiyin Xu, Yougen Zhou, Yihan Li, JingYuan Zhao, Qin Chen, Ningning Zhou, Kai Chen, Liang He
| Challenge: | Existing models focus on a single therapy, but complex cases require flexible strategies among various therapies. |
| Approach: | They propose a multi-session, multi-therapy, and highly realistic benchmark . it is designed to address three key challenges: 1) can we train a highly realistic AI counselor? 2) How to systematically evaluate an AI counselor?" |
| Outcome: | The proposed benchmark is annotated with extensive professional skills and includes over 677 meta-skills and 4577 atomic skills. |
Investigating and Mitigating the Multimodal Hallucination Snowballing in Large Vision-Language Models (2024.acl-long)
Copied to clipboard
Weihong Zhong, Xiaocheng Feng, Liang Zhao, Qiming Li, Lei Huang, Yuxuan Gu, Weitao Ma, Yuan Xu, Bing Qin
| Challenge: | Large Vision-Language Models (LVLMs) suffer from multimodal hallucinations . however, the generated hallucines could influence the models’ subsequent generation . |
| Approach: | They propose a framework to evaluate LVLMs' behaviors when encountering generated hallucinations and a method to revise the output distribution of LVLs with the one derived from the residual visual input. |
| Outcome: | The proposed framework reduces the performance of open-source LVLMs by 31%, indicating that they are prone to accept the generated hallucinations and make false claims that they would not have supported without distractions. |
Language Resource Efficient Learning for Captioning (2021.findings-emnlp)
Copied to clipboard
| Challenge: | XE loss and SC loss are both considered to be performance degradations for captioning tasks. |
| Approach: | They propose to generalize the single pairwise comparison in SC loss and use multiple generalized pairwise compares to reduce noise in baseline. |
| Outcome: | The proposed method outperforms state-of-the-art models on a video caption dataset using only half of the language resources. |
Toward Secure Tuning: Mitigating Security Risks from Instruction Fine-Tuning (2026.acl-long)
Copied to clipboard
Yanrui Du, Fenglei Fan, Sendong Zhao, Jiawei Cao, Ming Ma, Danyang Zhao, Shuren Qi, Ting Liu, Bing Qin
| Challenge: | Instruction Fine-Tuning (IFT) has emerged as a critical technique for customizing Large Language Models (LLMs) however, recent studies have revealed that IFT can compromise the built-in security mechanisms of LLMs, posing significant security risks. |
| Approach: | They propose a method that shifts learning burden onto security-robust parameters and propose 'warm-up' phase that preferentially trains Mods_Rob to learn low-level features with minimal security risk. |
| Outcome: | The proposed method reduces security risks without sacrificing performance gains across knowledge-intensive datasets. |
MuCDN: Mutual Conversational Detachment Network for Emotion Recognition in Multi-Party Conversations (2022.coling-1)
Copied to clipboard
| Challenge: | Emotion recognition in multi-party conversations is a challenging task that predicts the emotion for each utterance. |
| Approach: | They propose to separate conversations into detached threads to capture emotional clues in conversational context . they propose to use mutual detachment networks to perform context and speaker-specific modeling within detached thread. |
| Outcome: | The proposed model outperforms baseline models on two datasets. |
Don’t Lose Yourself! Empathetic Response Generation via Explicit Self-Other Awareness (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing attempts to generate empathy with other-awareness ignore to include self-a awareness to consider the own views of the self in their responses. |
| Approach: | They propose to include self-awareness to consider the own views of the self in empathetic response generation by integrating three stages of self-other awareness into the process. |
| Outcome: | The proposed method is superior to existing methods on the benchmark dataset. |
GainRAG: Preference Alignment in Retrieval-Augmented Generation through Gain Signal Synthesis (2025.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to retrieve information from large language models (LLMs) but they fail to address the preference gap between retrievers and LLMs. |
| Approach: | They propose a retrieval module that dynamically injects retrieved information into the input context of large language models (LLMs) This approach aligns the retriever’s and LLM’s preferences by defining a new metric, “gain”, which measure how well an input passage contributes to correct outputs. |
| Outcome: | The proposed approach has shown significant success in various NLP tasks, but there is a preference gap between retrievers and LLMs. |
Pruning via Merging: Compressing LLMs via Manifold Alignment Based Layer Merging (2024.emnlp-main)
Copied to clipboard
Deyuan Liu, Zhanyue Qin, Hairu Wang, Zhao Yang, Zecheng Wang, Fangying Rong, Qingbin Liu, Yanchao Hao, Bo Li, Xi Chen, Cunhang Fan, Zhao Lv, Dianhui Chu, Zhiying Tu, Dianbo Sui
| Challenge: | Existing methods for parameter pruning fail to utilize the knowledge from pruned parameters. |
| Approach: | They propose a method that uses manifold learning and the Information Bottleneck measure to merge similar layers to preserve model performance. |
| Outcome: | The proposed method outperforms pruning methods on multiple datasets and LLMs with quantization and achieves substantial compression ratios. |
CHAmbi: A New Benchmark on Chinese Ambiguity Challenges for Large Language Models (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Ambiguity is an inherent feature of language, whose management is crucial for effective communication and collaboration. |
| Approach: | They propose a dataset to evaluate LLMs' ability to handle ambiguity in Chinese by using a specialized Chinese multi-label disambiguation dataset formatted in Natural Language Inference. |
| Outcome: | The CHAmbi dataset comprises 4,991 pairs of premises and hypotheses, including 824 examples featuring a wide range of ambiguities. |
Data Augmentation using LLMs: Data Perspectives, Learning Paradigms and Challenges (2024.findings-acl)
Copied to clipboard
Bosheng Ding, Chengwei Qin, Ruochen Zhao, Tianze Luo, Xinze Li, Guizhen Chen, Wenhan Xia, Junjie Hu, Anh Tuan Luu, Shafiq Joty
| Challenge: | Data augmentation (DA) is a key technique for enhancing model performance by diversifying training examples without the need for additional data collection. |
| Approach: | They examine various strategies that utilize LLMs for data augmentation, including a novel exploration of learning paradigms where LLM-generated data is used for diverse forms of further training. |
| Outcome: | The proposed approach addresses the primary open challenges faced by LLMs in the field of large language models and aims to serve as a comprehensive guide for researchers and practitioners. |
ChildTalk: A Multi-Dialect Chinese Child Speech Corpus with Full-Length Child–Caregiver Conversations for Speech Recognition (2026.findings-acl)
Copied to clipboard
| Challenge: | Automatic speech recognition (ASR) for children remains challenging due to developmental variability and the scarcity of high-quality corpora. |
| Approach: | They propose a large-scale Chinese child speech corpus that contains 112.5 hours of speech from 498 children and 500 caregivers. |
| Outcome: | The proposed model improves in-domain and cross-domain performance on children's speech. |
A Versatile Adaptive Curriculum Learning Framework for Task-oriented Dialogue Policy Learning (2022.findings-naacl)
Copied to clipboard
| Challenge: | Existing training paradigms for dialogue policy learning with brute-force random sampling are expensive and lack reliable evaluation of difficulty scores. |
| Approach: | They propose a flexible adaptive curriculum learning framework that integrates curriculum learning with a generic global curriculum. |
| Outcome: | The proposed framework improves learning performance and efficiency on three public dialogue datasets. |
Separate the Wheat from the Chaff: A Post-Hoc Approach to Safety Re-Alignment for Fine-Tuned Language Models (2025.findings-acl)
Copied to clipboard
| Challenge: | Large language models achieve effective safety alignment at the time of release, but fine-tuning often compromises safety mechanisms. |
| Approach: | They propose a method that performs safety realignment for large language models . they identify unsafe delta parameters from the fine-tuned models and recalibrate the retained parameters . |
| Outcome: | The proposed method improves safety performance on safety benchmarks and jailbreak attacks while maintaining their performance on downstream tasks. |
Probing and Boosting Large Language Models Capabilities via Attention Heads (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to identifying capabilities rely on external signals with limited structural grounding . emergence of specific capabilities remains poorly understood . |
| Approach: | They propose a lightweight approach that links LLM capabilities to internal components by identifying correspondences at the level of attention heads. |
| Outcome: | The proposed approach improves accuracy on MMLU and BBH by 1 to 1.5 points over gradient-based method and 5 to 6 points over other intermediate-state baselines. |
Face-Sensitive Image-to-Emotional-Text Cross-modal Translation for Multimodal Aspect-based Sentiment Analysis (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing models focus on utilizing semantic information in the image but ignore using visual emotional cues. |
| Approach: | They propose a face-sensitive image-to-emotional-text translation method that captures visual emotional cues through facial expressions and selectively matches and fuses with the textual content. |
| Outcome: | The proposed method achieves state-of-the-art results on the Twitter-2015 and Twitter-2017 datasets. |
Can Large Language Models Understand You Better? An MBTI Personality Detection Dataset Aligned with Population Traits (2025.coling-main)
Copied to clipboard
Bohan Li, Jiannan Guan, Longxu Dou, Yunlong Feng, Dingzirui Wang, Yang Xu, Enbo Wang, Qiguang Chen, Bichen Wang, Xiao Xu, Yimeng Zhang, Libo Qin, Yanyan Zhao, Qingfu Zhu, Wanxiang Che
| Challenge: | Existing data on MBTI personality detection are based on self-reported labels and fail to capture the full range of population personality traits. |
| Approach: | They construct a manually annotated MBTI personality detection dataset with soft labels under the guidance of psychologists and use them to identify the task. |
| Outcome: | The MBTIBench is the first manually annotated MBti personality detection dataset with soft labels under the guidance of psychologists. |
C2D2 Dataset: A Resource for the Cognitive Distortion Analysis and Its Impact on Mental Health (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Cognitive distortions refer to patterns of irrational thinking that can lead to distorted perceptions of reality and mental health problems in individuals. |
| Approach: | They propose to use the C2D2 dataset to detect cognitive distortions in everyday life scenes to improve existing models of mental health detection. |
| Outcome: | The proposed dataset contains 7,500 cognitive distortion thoughts in everyday life scenes. |
RU22Fact: Optimizing Evidence for Multilingual Explainable Fact-Checking on Russia-Ukraine Conflict (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods to verify factuality of claims do not provide sufficient evidence for explainable fact-checking systems. |
| Approach: | They propose a method to automatically retrieve and summarize evidence from the Web and a novel multilingual explainable fact-checking dataset on the Russia-Ukraine conflict in 2022. |
| Outcome: | The proposed method can retrieve and summarize evidence from the Web and generate explanations in 16 languages. |
A Text-Centered Shared-Private Framework via Cross-Modal Prediction for Multimodal Sentiment Analysis (2021.findings-acl)
Copied to clipboard
| Challenge: | Existing studies treat all three modal features equally and implicitly explore the interactions between different modalities. |
| Approach: | They propose a text-centered shared-private framework for multimodal fusion . they propose modalities that can provide shared and private semantics . |
| Outcome: | The proposed framework outperforms baselines on the MOSEI and MOSI datasets. |
AssoCiAm: A Benchmark for Evaluating Association Thinking while Circumventing Ambiguity (2025.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in multimodal large language models (MLLMs) have garnered significant attention, offering a promising pathway toward artificial general intelligence (AGI). |
| Approach: | They propose a benchmark to evaluate associative ability while circumventing the inherent ambiguity in association tasks by decomposing ambiguities into two types and propose 'assoCiAm' they conduct extensive experiments on MLLMs, revealing a strong positive correlation between cognition and association. |
| Outcome: | The proposed method shows that ambiguity in association evaluations makes MLLMs more random-like and the model's behavior more random. |
Question Tells You Where the Answer Is: Intention-aware Long-Context KV Cache Compression (2026.acl-long)
Copied to clipboard
Liang Zhao, Xiaocheng Feng, Weihong Zhong, Lei Huang, Kun Zhu, Baoxin Wang, Dayong Wu, Guoping Hu, Ting Liu, Bing Qin
| Challenge: | Recent methods to reduce the KV cache size fail to identify crucial KVs for generation while excluding others accurately, resulting in severe information loss. |
| Approach: | They propose an intention-aware KV cache eviction method that identifies and retains crucial KVs according to the attention distribution of intention, which semantically reflects the user’s goal and determines which part of the context is relevant. |
| Outcome: | The proposed method can maintain the model performance while reducing the KV cache size from 128K to 2K, leading to a 6.3x increase in decoding speed and 7.8x enhancement in memory efficiency compared to the default setting. |
KG-RAG: Enhancing GUI Agent Decision-Making via Knowledge Graph-Driven Retrieval-Augmented Generation (2025.emnlp-main)
Copied to clipboard
Ziyi Guan, Jason Chun Lok Li, Zhijian Hou, Pingping Zhang, Donglai Xu, Yuzhi Zhao, Mengyang Wu, Jinpeng Chen, Thanh-Toan Nguyen, Pengfei Xian, Wenao Ma, Shengchao Qin, Graziano Chesi, Ngai Wong
| Challenge: | Recent advances in GUI agents have limited app-specific knowledge of complex mobile tasks. |
| Approach: | They propose a Knowledge Graph-driven Retrieval-Augmented Generation framework that transforms fragmented UTGs into structured vector databases for efficient real-time retrieval. |
| Outcome: | The proposed framework outperforms existing methods in a 75.8% success rate and 84.6% decision accuracy test across mobile apps. |
PromptRank: Unsupervised Keyphrase Extraction Using Prompt (2023.acl-long)
Copied to clipboard
| Challenge: | Existing keyphrase extraction methods struggle with document and candidate length discrepancies or fail to fully utilize the pre-trained language model without further fine-tuning. |
| Approach: | They propose an unsupervised keyphrase extraction approach that uses a pre-trained language model to rank candidates based on document embeddings. |
| Outcome: | The proposed approach outperforms the existing keyphrase extraction approach on six benchmarks. |
Investigating Ensemble Methods for Model Robustness Improvement of Text Classifiers (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to reduce model's reliance on bias features ignore the learnability of these features. |
| Approach: | They propose to reduce models' reliance on bias features by first training models with fixed low-capacity models which ignore the learnability of the bias features. |
| Outcome: | The proposed models can perform better on out-of-distribution datasets than baseline models with a more sophisticated model design. |
LiPO: Listwise Preference Optimization through Learning-to-Rank (2025.naacl-long)
Copied to clipboard
Tianqi Liu, Zhen Qin, Junru Wu, Jiaming Shen, Misha Khalman, Rishabh Joshi, Yao Zhao, Mohammad Saleh, Simon Baumgartner, Jialu Liu, Peter J Liu, Xuanhui Wang
| Challenge: | Recent work on language models with curated feedback provides promising alternatives to RLHF . multiple responses can be ranked by reward models or AI feedback, but there is no study on directly fitting upon a list of responses. |
| Approach: | They propose a method that aligns language models with curated human feedback . they propose SLiC and DPO as promising alternatives to traditional RLHF . |
| Outcome: | The proposed method outperforms DPO and SLiC on several preference alignment tasks with curated and real rankwise preference data. |
SDPO: Segment-Level Direct Preference Optimization for Social Agents (2025.acl-long)
Copied to clipboard
Aobo Kong, Wentao Ma, Shiwan Zhao, Yongbin Li, Yuchuan Wu, Ke Wang, Xiaoqian Liu, Qicheng Li, Yong Qin, Fei Huang
| Challenge: | Direct Preference Optimization (DPO) has proven effective in aligning LLM behavior with human preferences across various tasks, but is limited in multi-turn social interactions. |
| Approach: | They propose a method which dynamically selects key segments within interactions to optimize multi-turn agent behavior. |
| Outcome: | The proposed methods outperform existing methods and proprietary LLMs on the SOTOPIA benchmark and show that they can improve social intelligence. |
Chain of Strategy Optimization Makes Large Language Models Better Emotional Supporter (2025.findings-emnlp)
Copied to clipboard
Weixiang Zhao, Xingyu Sui, Xinyang Han, Yang Deng, Yulin Hu, Jiahe Guo, Libo Qin, Qianyun Du, Shijin Wang, Yanyan Zhao, Bing Qin, Ting Liu
| Challenge: | Existing supervised fine-tuning (SFT) fails to address these issues, as it trains models on single gold-standard responses without modeling nuanced strategy trade-offs. |
| Approach: | They propose a two-stage framework that optimizes strategy selection preferences at each dialogue turn. |
| Outcome: | The proposed framework improves strategy selection preferences at each dialogue turn. |
GUICourse: From General Vision Language Model to Versatile GUI Agent (2025.acl-long)
Copied to clipboard
Wentong Chen, Junbo Cui, Jinyi Hu, Yujia Qin, Junjie Fang, Yue Zhao, Chongyi Wang, Jun Liu, Guirong Chen, Yupeng Huo, Yuan Yao, Yankai Lin, Zhiyuan Liu, Maosong Sun
| Challenge: | Graphical User Interfaces (GUIs) are a pivotal medium for human-computer interaction. |
| Approach: | They propose a series of datasets for training visual-based GUI agents using general VLMs. |
| Outcome: | The proposed GUICourse datasets show that even a small-sized GUI agent performs better on GUI tasks. |
Bridging Reasoning and Action: Hybrid LLM–RL Framework for Efficient Cross-Domain Task-Oriented Dialogue (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to solve cross-domain task-oriented dialogues are brittle when cross- domain constraints are not directly grounded in surface text or require commonsense inference. |
| Approach: | They propose a framework that makes LLM-derived constraint reasoning usable for RL. |
| Outcome: | Experiments show that the proposed framework outperforms single-model baselines on long-horizon tasks. |
Natural Logic at the Core: Dynamic Rewards for Entailment Tree Generation (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing approaches to generating entailment trees lack logical consistency . static reward structures or intricate dependencies within multi-step reasoning are often ignored . |
| Approach: | They propose a method that integrates natural logic principles into reinforcement learning to guide entailment tree generation. |
| Outcome: | Experiments on EntailmentBank show that the proposed method improves interpretability and generalization. |
Length Extrapolation of Transformers: A Survey from the Perspective of Positional Encoding (2024.findings-emnlp)
Copied to clipboard
Liang Zhao, Xiachong Feng, Xiaocheng Feng, Weihong Zhong, Dongliang Xu, Qing Yang, Hongtao Liu, Bing Qin, Ting Liu
| Challenge: | Existing methods to enhance length extrapolation of large language models have been developed, but a systematic survey is lacking. |
| Approach: | They propose to examine the effects of positional encoding on length extrapolation. |
| Outcome: | The proposed methods improve the extrapolation of large language models, but they are still lacking a systematic survey. |
ENPMR-Bench: Benchmarking Proactive Memory Retrieval for Emotional Support Agents (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing research treats memory as a mechanism for factual retention, neglecting its role in shaping users’ emotional experiences. |
| Approach: | They propose a benchmark for evaluating Emotional Need-aware Proactive Memory Retrieval (ENPMR) it enables agents to infer users’ latent emotional needs and proactively retrieve appropriate memories to support empathetic interaction. |
| Outcome: | The proposed benchmark includes over 1,800 memory-augmented dialogues and defines structured mappings between emotional needs and supportive memory types. |
M3ED: Multi-modal Multi-scene Multi-label Emotional Dialogue Database (2022.acl-long)
Copied to clipboard
| Challenge: | Existing data resources to support multimodal affective analysis in dialogues are limited in scale and diversity. |
| Approach: | They propose a multimodal multi-scene multi-label Emotional Dialogue dataset, M3ED, which contains 990 dyadic emotional dialogues from 56 different TV series. |
| Outcome: | The proposed dataset contains 990 dyadic emotional dialogues from 56 different TV series, a total of 9,082 turns and 24,449 utterances. |
LLMs May Perform MCQA by Selecting the Least Incorrect Option (2025.coling-main)
Copied to clipboard
| Challenge: | Multiple Choice Question Answering (MCQA) is a fundamental format for various tasks in NLP, such as commonsense reasoning. |
| Approach: | They propose a method to increase the number of correct options in a dataset. |
| Outcome: | The proposed method improves the performance of multiple choice question answering (MCQA) and improves its accuracy. |
Look Beyond Feeling: Unveiling Latent Needs from Implicit Expressions for Proactive Emotional Support (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) are gaining popularity as scalable tools for mental health support . however, nearly half of individuals do not receive timely support due to limited selfawareness or reluctance to seek help. |
| Approach: | They propose a proactive emotional support framework that leverages principles of active listening to uncover implicit user needs. |
| Outcome: | The proposed model elicits implicit emotional needs and delivers empathetic support compared to baselines . |
Distilling Knowledge for Search-based Structured Prediction (P18-1)
Copied to clipboard
| Challenge: | Existing studies have focused on the performance of structured prediction models, but they are often limited by the ambiguities of the reference policy. |
| Approach: | They propose to distill an ensemble of multiple models trained with different initializations into a single model and use it to explore the search space. |
| Outcome: | The proposed model outperforms the greedy models on two typical search-based structured prediction tasks and achieves 1.32 in LAS and 2.65 in BLEU over strong baselines. |
Self-Evolving GPT: A Lifelong Autonomous Experiential Learner (2024.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to provide LLMs with textual task-solving experience rely on manual efforts to acquire and apply such experience for each task. |
| Approach: | They propose a lifelong autonomous experiential learning framework based on LLMs that learns and accumulates experience through experience transfer and induction. |
| Outcome: | The proposed framework performs reliably in each intermediate step and improves GPT-3.5 and GPT-4 on widely used NLP datasets. |
ESDM: Early Sensing Depression Model in Social Media Streams (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing approaches to use social media data for depression detection are based on traditional risk detection (TRD) and early risk detection of depression (ERD). |
| Approach: | They propose a model that uses two modules: classification with partial information module (CPI) and decision for classification moment module (DMC) and an early detection loss function. |
| Outcome: | The proposed model outperforms benchmarks in both accuracy and accuracy with evolving partial data. |
STINMatch: Semi-Supervised Semantic-Topological Iteration Network for Financial Risk Detection via News Label Diffusion (2023.emnlp-main)
Copied to clipboard
Xurui Li, Yue Qin, Rui Zhu, Tianqianjin Lin, Yongming Fan, Yangyang Kang, Kaisong Song, Fubang Zhao, Changlong Sun, Haixu Tang, Xiaozhong Liu
| Challenge: | Commercial news provides rich semantics and timely information for automated financial risk detection. |
| Approach: | They propose a semi-supervised Semantic-Topological Iteration Network, STINMatch, along with a news-enterprise knowledge graph to endorse the risk detection enhancement. |
| Outcome: | The proposed model outperforms existing models in terms of generalization and semantics and annotation. |
AdaSteer: Your Aligned LLM is Inherently an Adaptive Jailbreak Defender (2025.emnlp-main)
Copied to clipboard
Weixiang Zhao, Jiahe Guo, Yulin Hu, Yang Deng, An Zhang, Xingyu Sui, Xinyang Han, Yanyan Zhao, Bing Qin, Tat-Seng Chua, Ting Liu
| Challenge: | Activation steering offers training-free defense but relies on fixed steering coefficients, resulting in suboptimal protection and increased false rejections of benign inputs. |
| Approach: | They propose an adaptive activation steering method that dynamically adjusts model behavior based on input characteristics. |
| Outcome: | The proposed method outperforms baseline methods across multiple jailbreak attacks with minimal impact on utility. |
End-to-End Learnable Psychiatric Scale Guided Risky Post Screening for Depression Detection on Social Media (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to detect depression from social media posting history are limited by frozen screening models and lack of learning. |
| Approach: | They propose to use a frozen screening model to train a risky post detection model with psychiatric scales to enable a learnable end-to-end learning process. |
| Outcome: | The proposed model outperforms several strong baseline methods and qualitative analysis confirms that it better captures users’ mental states than others. |
DORA: A Dual-Objective Reinforcement Learning Framework for Effective and Efficient Multimodal Agentic Search (2026.acl-long)
Copied to clipboard
Guangming Qin, Yuhao Deng, Yukun Zhao, Zhenyang Li, Junfeng Wang, Dawei Yin, Ye Yuan, Guoren Wang, Yizhou Yan, Chengliang Chai, Lei Cao
| Challenge: | Existing methods to train large language models overlook quality of intermediate search results . existing methods often invoke search calls during reasoning, making inference inefficient . |
| Approach: | They propose a dual-objective reinforcement learning framework to improve search strategies of MLLMs . DORA outperforms state-of-the-art methods, achieving up to 8.4% higher accuracy . |
| Outcome: | The proposed model outperforms state-of-the-art methods while reducing search calls by 9.7%. |
Less Is More: Domain Adaptation with Lottery Ticket for Reading Comprehension (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Existing domain adaptation paradigms for reading comprehension require large amounts of annotation data to achieve the desired task performance. |
| Approach: | They propose a few-shot domain adaptation paradigm for reading comprehension . they introduce self-attention attribution to weigh parameters and refine the lottery subnetwork . |
| Outcome: | The proposed model outperforms the full model fine-tuning adaptation on four out of five domains with a small amount of data available for adaptation. |
Learning to Initialize: Can Meta Learning Improve Cross-task Generalization in Prompt Tuning? (2023.acl-long)
Copied to clipboard
| Challenge: | Prompt tuning (PT) based on frozen pre-trained language models has shown remarkable performance in few-shot learning . however, it relies heavily on good initialization of the prompt embeddings. |
| Approach: | They propose to use meta prompt tuning to improve cross-task generalization by learning to initialize prompt embeddings from other relevant tasks. |
| Outcome: | The proposed method outperforms PT on classification tasks, but not multi-task learning. |
Understanding Large Language Model Vulnerabilities to Social Bias Attacks (2025.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated remarkable linguistic capabilities across tasks . however, there is a growing concern about their potential to perpetuate social biases . |
| Approach: | They evaluate LLMs across gender, racial, and religious bias types . they also explore cross-bias and multiple-biases attacks . |
| Outcome: | The proposed models are more susceptible to gender bias attacks than racial or religious biases. |
Multi-Input Multi-Output Sequence Labeling for Joint Extraction of Fact and Condition Tuples from Scientific Text (D19-1)
Copied to clipboard
| Challenge: | Existing methods to extract factual tuples from scientific text do not consider conditions. |
| Approach: | They propose a new sequence labeling framework to jointly extract fact and condition tuples from scientific sentences. |
| Outcome: | The proposed framework improves F1 score relative to existing methods by 4.2% and 6.2% on bioNLP2013. |
Better Zero-Shot Reasoning with Role-Play Prompting (2024.naacl-long)
Copied to clipboard
Aobo Kong, Shiwan Zhao, Hao Chen, Qicheng Li, Yong Qin, Ruiqi Sun, Xin Zhou, Enzhi Wang, Xiaohang Dong
| Challenge: | Recent years have witnessed a paradigm shift in natural language processing, driven by large language models such as GPT-3, PaLM, and Llama. |
| Approach: | They propose a strategy for role-play prompting and assess its performance under the zero-shot setting. |
| Outcome: | The proposed method outperforms the standard zero-shot prompting approach across 12 reasoning benchmarks. |
Deciphering the Impact of Pretraining Data on Large Language Models through Machine Unlearning (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing studies have suggested that the composition of the pretraining corpus exerts a significant impact upon the performance of LLMs. |
| Approach: | They analyze the impact of 48 datasets from 5 major categories of pretraining data of Large Language Models and measure their impacts on LLMs using benchmarks about nine major categories. |
| Outcome: | The proposed analysis provides insights into the organization of data to support more efficient pretraining of Large Language Models. |
TransESC: Smoothing Emotional Support Conversation via Turn-Level State Transition (2023.findings-acl)
Copied to clipboard
| Challenge: | Emotion Support Conversation (ESC) is a goaldirected task with the goal of reducing the emotional distress of people. |
| Approach: | They propose to take turn-level state Transitions of ESC from three perspectives to generate smooth transitions between utterances. |
| Outcome: | The proposed method generates smoother and more effective responses on automatic and human evaluations. |
Progressive Self-Training with Discriminator for Aspect Term Extraction (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to extract aspect terms from review sentences are limited due to lack of annotated data. |
| Approach: | They propose to refine conventional self-training to progressive self-teaching to reduce noise . they use a discriminator to filter the noisy pseudo-labels. |
| Outcome: | The proposed model outperforms baseline models and achieves state-of-the-art performance on four SemEval datasets. |
Analyzing the Rapid Generalization of SFT via the Perspective of Attention Head Activation Patterns (2025.acl-long)
Copied to clipboard
| Challenge: | Currently, LLMs learn in a data-driven schema while the instructions about complex tasks are both scarce and hard to collect or construct. |
| Approach: | They employ a gradient-based method to dissect the process that the Supervised Fine-tuning Process (SFT) adapts LLMs to downstream tasks via the perspective of attention patterns. |
| Outcome: | The proposed method dissects the process that the SFT process adapts LLMs to downstream tasks via the perspective of attention patterns. |
WISCA: A Lightweight Model Transition Method to Improve LLM Training via Weight Scaling (2026.findings-acl)
Copied to clipboard
Jiacheng Li, Jianchao Tan, Zhidong Yang, Pingwei Sun, Feiye Huo, Jiayu Qin, Xiangyu Zhang, Maoxin He, Guangming Tan, Weile Jia, Xunliang Cai, Tong Zhao
| Challenge: | Recent advances in training optimization for Transformer-based large language models lack systematic optimization of weight patterns during training. |
| Approach: | They propose a Weight Scaling method that rescales weights while preserving model outputs to improve model training efficiency and model quality. |
| Outcome: | The proposed method significantly improves convergence quality and loss reduction in LLMs with Grouped Query Attention architectures and LoRA fine-tuning tasks. |
ESCoT: Towards Interpretable Emotional Support Dialogue Systems (2024.acl-long)
Copied to clipboard
| Challenge: | Emotion-focused and strategy-driven chain-of-thought (ESCoT) is a new paradigm for emotional support dialogues. |
| Approach: | They propose an emotional support response generation scheme to improve interpretability . they generate a dataset and develop a model to generate dialogue responses with better interpretability. |
| Outcome: | The proposed scheme can generate dialogue responses with better interpretability. |
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. |
UNO Arena for Evaluating Sequential Decision-Making Capability of Large Language Models (2024.emnlp-main)
Copied to clipboard
Zhanyue Qin, Haochuan Wang, Deyuan Liu, Ziyang Song, Cunhang Fan, Zhao Lv, Jinlin Wu, Zhen Lei, Zhiying Tu, Dianhui Chu, Xiaoyan Yu, Dianbo Sui
| Challenge: | Existing LLMs demonstrate powerful capabilities between tasks, but can they make sequential decisions? |
| Approach: | They propose to evaluate sequential decision-making capability of large language models (LLMs) using novel metrics based Monte Carlo methods. |
| Outcome: | The proposed benchmark improves sequential decision-making performance compared to the vanilla LLM player. |
VisBias: Measuring Explicit and Implicit Social Biases in Vision Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Identifying and addressing potential social biases is essential to prevent harm to users. |
| Approach: | They examine explicit and implicit biases exhibited by Vision-Language Models . they pose questions related to gender and racial differences to test their models . |
| Outcome: | The proposed models are used in image description tasks, form completion tasks and medical applications. |
Lifelong Event Detection with Embedding Space Separation and Compaction (2024.naacl-short)
Copied to clipboard
| Challenge: | Existing methods for event detection are prone to forgetting due to overlap between memory data and the previously learned embedding space. |
| Approach: | They propose a method that embeds feature distributions away from the previous embedding space and mitigates overfitting by a memory calibration mechanism. |
| Outcome: | The proposed method outperforms existing state-of-the-art methods with extensive experiments. |
Psychological Counseling Cannot Be Achieved Overnight: Automated Psychological Counseling Through Multi-Session Conversations (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing studies on single-session counseling are limited to a single-session setting. |
| Approach: | They propose to use a large language model to deliver automated psychological counseling to a dataset constructed using real client profiles from publicly available psychological case reports. |
| Outcome: | The proposed model performs better than baseline models across multiple sessions. |
LLMBox: A Comprehensive Library for Large Language Models (2024.acl-demos)
Copied to clipboard
Tianyi Tang, Hu Yiwen, Bingqian Li, Wenyang Luo, ZiJing Qin, Haoxiang Sun, Jiapeng Wang, Shiyi Xu, Xiaoxue Cheng, Geyang Guo, Han Peng, Bowen Zheng, Yiru Tang, Yingqian Min, Yushuo Chen, Jie Chen, Ranchi Zhao, Luran Ding, Yuhao Wang, Zican Dong, Xia Chunxuan, Junyi Li, Kun Zhou, Xin Zhao, Ji-Rong Wen
| Challenge: | a library to facilitate the development, use, and evaluation of large language models (LLMs) is presented. |
| Approach: | They propose a unified library to facilitate the development, use and evaluation of large language models (LLMs). |
| Outcome: | The proposed library is based on extensive experiments in a variety of evaluation settings. |
LESA: Learnable LLM Layer Scaling-Up (2025.acl-long)
Copied to clipboard
| Challenge: | Existing methods for depth scaling-up rely on empirical heuristic rules for layer duplication, resulting in poor initialization and slower convergence during continual pre-training. |
| Approach: | They propose a method for learning latent parameters between layers by concatenating parameters from each layer and applying Singular Value Decomposition. |
| Outcome: | Experiments show that LESA outperforms baseline models with less than half the cost of existing methods. |
MessToClean: Evidence-Grounded Structure-Preserving Reconstruction for Real-World Degraded Exam Paper Images (2026.acl-long)
Copied to clipboard
Jiayi Tuo, Cheng Tang, Zihan Wang, Chenyue Zhou, Yao Li, Yanbiao Ma, Chao Wang, Wei Dai, Mingxuan Wang, Shitong Qin, Ziwei Zhao
| Challenge: | Existing Multimodal Large Language Models (MLLMs) fail under RDEI, leading to disrupted structure and evidence-unsupported hallucinations. |
| Approach: | They propose a backbone-agnostic, evidence-driven pipeline that treats off-the-shelf MLLMs as interchangeable components to improve stem consistency and figure consistency. |
| Outcome: | The proposed pipeline improves stem consistency by 1.01-3.18%, figure consistency by 0.50-49.16%, and refusal F1 by 1.06-10.88% across question types. |
Large Language Models Are Still Misled by Simple Bias Ensembles (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing benchmarks for large language models are constrained to datasets where each sample is manually injected with only one type of bias. |
| Approach: | They propose a multi-bias benchmark where each sample contains multiple types of biases. |
| Outcome: | The proposed benchmark shows that existing LLMs and debiasing methods perform poorly on this benchmark, highlighting the challenge of eliminating compounded biases. |
TEA-Bench: A Systematic Benchmarking of Tool-enhanced Emotional Support Dialogue Agent (2026.acl-long)
Copied to clipboard
| Challenge: | Existing ESC systems focus on affective support in text-only settings, ignoring how external tools can enable factual grounding and reduce hallucination in multi-turn emotional support. |
| Approach: | They propose a benchmark for evaluating tool-augmented agents in ESC with realistic emotional scenarios and an MCP-style tool environment. |
| Outcome: | The proposed benchmarks show that tool augmentation improves emotional support quality and reduces hallucination, but weaker models benefit only marginally. |
SATQuest: A Verifier for Logical Reasoning Evaluation and Reinforcement Fine-Tuning of LLMs (2026.acl-long)
Copied to clipboard
Yanxiao Zhao, Yaqian Li, Zi-Hao Bo, Rinyoichi Takezoe, Haojia Hui, Mo Guang, null Renlei, Xiaolin Qin, Kaiwen Long
| Challenge: | Large language models exhibit strong general reasoning abilities, yet the community lacks controllable, scalable, and verifiable tools to analyze and improve them. |
| Approach: | They propose a verifier that generates diverse SAT-based reasoning tasks from CNF instances and checks answers objectively with PySAT. |
| Outcome: | The proposed verifier generates diverse SAT-based reasoning tasks from CNF instances and checks answers objectively with PySAT. |
Missing Modality Imagination Network for Emotion Recognition with Uncertain Missing Modalities (2021.acl-long)
Copied to clipboard
| Challenge: | Existing multimodal fusion models trained on full-modality samples fail when partial modalities are missing. |
| Approach: | They propose a model to deal with the uncertain missing modality problem by learning robust joint multimodal representations that can predict the representation of any missing modal given available modalities under different missing-modality conditions. |
| Outcome: | The proposed model significantly improves performance under uncertain missing-modality testing conditions and full-modalities ideal testing conditions. |
Balancing Forget Quality and Model Utility: A Reverse KL-Divergence Knowledge Distillation Approach for Better Unlearning in LLMs (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing methods for unlearning large language models struggle with forget quality and model utility, leading to over-unlearning or partial unlearning. |
| Approach: | They propose a method that uses reverse KL-divergence based knowledge distillation for unlearning to achieve significant forget quality while maintaining model utility. |
| Outcome: | The proposed method outperforms existing methods in forget quality and model utility with larger unlearning datasets. |
Advancing Large Language Model Attribution through Self-Improving (2024.emnlp-main)
Copied to clipboard
Lei Huang, Xiaocheng Feng, Weitao Ma, Liang Zhao, Yuchun Fan, Weihong Zhong, Dongliang Xu, Qing Yang, Hongtao Liu, Bing Qin
| Challenge: | Teaching large language models to generate text with citations to evidence sources requires high-quality attribution data, which is costly and labor-intensive. |
| Approach: | They propose a framework for iteratively improving the attribution capability of large language models (LLMs) by attributing output to verifiable sources. |
| Outcome: | Experiments on three open-domain question-answering datasets show that START improves in aggregating information across multiple sources. |
Large Language Models as Foundations for Next-Gen Dense Retrieval: A Comprehensive Empirical Assessment (2024.emnlp-main)
Copied to clipboard
| Challenge: | Pre-trained language models have limited generalization capabilities and performance challenges. |
| Approach: | They evaluate 15 different backbone LLMs and non-LLMs to evaluate their performance . larger models and extensive pre-training consistently enhance in-domain accuracy and data efficiency . |
| Outcome: | The results show that larger models and extensive pre-training enhance in-domain accuracy and data efficiency. |
DialogueEIN: Emotion Interaction Network for Dialogue Affective Analysis (2022.coling-1)
Copied to clipboard
| Challenge: | Emotion Recognition in Conversation (ERC) has attracted increasing research attention in recent years. |
| Approach: | They propose to model the emotional interactions between speakers to simulate the emotional inertia, emotional stimulus, global and local emotional evolution in dialogues. |
| Outcome: | The proposed model can achieve superior performance compared to state-of-the-art methods on four ERC benchmark datasets, IEMOCAP, MELD, EmoryNLP and DailyDialog. |
TextFlint: Unified Multilingual Robustness Evaluation Toolkit for Natural Language Processing (2021.acl-demo)
Copied to clipboard
Xiao Wang, Qin Liu, Tao Gui, Qi Zhang, Yicheng Zou, Xin Zhou, Jiacheng Ye, Yongxin Zhang, Rui Zheng, Zexiong Pang, Qinzhuo Wu, Zhengyan Li, Chong Zhang, Ruotian Ma, Zichu Fei, Ruijian Cai, Jun Zhao, Xingwu Hu, Zhiheng Yan, Yiding Tan, Yuan Hu, Qiyuan Bian, Zhihua Liu, Shan Qin, Bolin Zhu, Xiaoyu Xing, Jinlan Fu, Yue Zhang, Minlong Peng, Xiaoqing Zheng, Yaqian Zhou, Zhongyu Wei, Xipeng Qiu, Xuanjing Huang
| Challenge: | Existing approaches to textual robustness evaluation focus on slightly modifying the input data, which maintains the original meaning and results in a different prediction. |
| Approach: | They propose a multilingual robustness evaluation toolkit for NLP that integrates universal text transformations, task-specific transformations and adversarial attack. |
| Outcome: | The toolkit includes universal text transformation, task-specific transformation, adversarial attack, subpopulation, and their combinations to provide comprehensive robustness analyses. |
Prompt Combines Paraphrase: Teaching Pre-trained Models to Understand Rare Biomedical Words (2022.coling-1)
Copied to clipboard
Haochun Wang, Chi Liu, Nuwa Xi, Sendong Zhao, Meizhi Ju, Shiwei Zhang, Ziheng Zhang, Yefeng Zheng, Bing Qin, Ting Liu
| Challenge: | Pre-trained models perform poorly with limited data and rare biomedical words. |
| Approach: | They propose to use prompt to fine-tune pre-trained models for biomedical domain tuning with a simple approach. |
| Outcome: | The proposed method achieves up to 6% improvement in biomedical natural language inference task without any extra parameters or training steps using few-shot vanilla prompt settings. |
SAPT: A Shared Attention Framework for Parameter-Efficient Continual Learning of Large Language Models (2024.acl-long)
Copied to clipboard
Weixiang Zhao, Shilong Wang, Yulin Hu, Yanyan Zhao, Bing Qin, Xuanyu Zhang, Qing Yang, Dongliang Xu, Wanxiang Che
| Challenge: | Existing methods to address catastrophic forgetting and knowledge transfer in large language models (LLMs) ignore potential of aligning the two modules to effectively address catastrophic forgetting and knowledge transfers simultaneously. |
| Approach: | They propose a Shared Attentive Learning & Selection module to align the PET learning and selection modules to address catastrophic forgetting and knowledge transfer simultaneously. |
| Outcome: | Experiments on two CL benchmarks show that the proposed framework is superior when scaled to different model sizes, different model architectures and unseen tasks. |
Infrared-LLaVA: Enhancing Understanding of Infrared Images in Multi-Modal Large Language Models (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for infrared modeling ignore supervisory signals of infra-modality-specific attributes, which may lead to biased understanding of in-frarea images. |
| Approach: | They propose a multi-agent generation system which transfers knowledge from visible images to generate infrared image-text pairs and infra-instructional data. |
| Outcome: | The proposed system generates infrared image-text pairs and infra-response data and is able to answer common infreas tasks with the proposed model. |
SpeechLLM-as-Judges: Towards General and Interpretable Speech Quality Evaluation (2026.acl-long)
Copied to clipboard
Hui Wang, Jinghua Zhao, Yifan Yang, Shujie Liu, Junyang Chen, Yanzhe Zhang, Shiwan Zhao, Jinyu Li, Jiaming Zhou, Haoqin Sun, Yan Lu, Yong Qin
| Challenge: | Existing methods for evaluating the perceptual quality of synthetic speech are limited due to the complexity of perceptual quality factors and the diversity of speech generation tasks. |
| Approach: | They propose a new paradigm for enabling large language models to conduct structured speech quality evaluation using a large-scale dataset. |
| Outcome: | The proposed model performs well across tasks and languages. |