Papers by Wen Zhao
Learning to Imagine: Visually-Augmented Natural Language Generation (2023.acl-long)
Copied to clipboard
| Challenge: | Existing methods for natural language generation are pre-trained on text-only corpora, resulting in visual commonsense. |
| Approach: | They propose a method that makes pre-trained language models learn to imagine for visually-augmented natural language generation. |
| Outcome: | The proposed method is compatible with Transformer-based architecture. |
Dual-Axis Generative Reward Model Toward Semantic and Turn-taking Robustness in Interactive Spoken Dialogue Models (2026.acl-long)
Copied to clipboard
Yifu Chen, Shengpeng Ji, Zhengqing Liu, Qian Chen, Wen Wang, Ziqing Wang, Yangzhuo Li, Tianle Liang, Zhou Zhao
| Challenge: | Reinforcement learning (RL) has improved text- and vision-language models, but its application in SDMs is hindered. |
| Approach: | They propose a dual-axis Generative Reward Model that provides semantic quality and interaction timing for SDMs. |
| Outcome: | The proposed model achieves state-of-the-art performance on interaction-quality assessment across a wide spectrum of datasets. |
There Are a Thousand Hamlets in a Thousand People’s Eyes: Enhancing Knowledge-grounded Dialogue with Personal Memory (2022.acl-long)
Copied to clipboard
| Challenge: | Existing methods for knowledge selection focus on relevance between knowledge and dialogue context, ignoring personal preference for knowledge. |
| Approach: | They propose to introduce personal memory into knowledge selection in chatbots to address personalization issue by integrating personal memory and inverse mapping into a closed loop. |
| Outcome: | The proposed method outperforms existing methods significantly on automatic evaluation and human evaluation. |
CodeScope: An Execution-based Multilingual Multitask Multidimensional Benchmark for Evaluating LLMs on Code Understanding and Generation (2024.acl-long)
Copied to clipboard
Weixiang Yan, Haitian Liu, Yunkun Wang, Yunzhe Li, Qian Chen, Wen Wang, Tingyu Lin, Weishan Zhao, Li Zhu, Hari Sundaram, Shuiguang Deng
| Challenge: | Existing benchmarks for evaluating the code understanding and generation capacities of Large Language Models are insufficient . existing benchmarks focus on a narrow range of popular programming languages and specific tasks . |
| Approach: | They propose an execution-based, multilingual, multitask evaluation benchmark for LLMs . they evaluate coding performance from three dimensions: length, difficulty, efficiency . |
| Outcome: | The proposed benchmark covers 43 programming languages and eight coding tasks. |
CORAL: Benchmarking Multi-turn Conversational Retrieval-Augmented Generation (2025.findings-naacl)
Copied to clipboard
Yiruo Cheng, Kelong Mao, Ziliang Zhao, Guanting Dong, Hongjin Qian, Yongkang Wu, Tetsuya Sakai, Ji-Rong Wen, Zhicheng Dou
| Challenge: | Existing research focuses on single-turn RAG, leaving a gap in addressing multi-turn conversations . a new benchmark is designed to assess RAG systems in realistic multi-turned conversations based on Wikipedia . |
| Approach: | They propose a large-scale benchmark to assess RAG systems in multi-turn contexts . CORAL includes diverse information-seeking conversations automatically derived from Wikipedia . authors propose unified framework to standardize various conversational RAG methods . |
| Outcome: | The proposed framework supports three core tasks of conversational RAG: passage retrieval, response generation, and citation labeling. |
Context-Tuning: Learning Contextualized Prompts for Natural Language Generation (2022.coling-1)
Copied to clipboard
| Challenge: | Recent studies have shown that pretrained language models (PLMs) lack sufficient consideration of input semantics to generate natural language. |
| Approach: | They propose a continuous prompting approach to fine-tune PLMs for natural language generation by modeling an inverse generation process from output to input. |
| Outcome: | The proposed method fine-tunes only 0.12% of the parameters while maintaining good performance. |
Rethinking Sample Polarity in Reinforcement Learning with Verifiable Rewards (2026.acl-long)
Copied to clipboard
| Challenge: | Large reasoning models are typically trained using reinforcement learning with verifiable reward (RLVR) positive and negative self-generated rollouts are used to update the model's policy . positive samples sharpen existing correct reasoning patterns, while negative samples encourage exploration of new reasoning paths. |
| Approach: | They propose a method that allocates advantage signals to key tokens across different polarities. |
| Outcome: | The proposed method improves the ability of large reasoning models to learn from their own generated rollouts. |
Enhancing Parameter-efficient Fine-tuning with Simple Calibration Based on Stable Rank (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods for lightweight fine-tuning are ineffective in low-resource settings but fail in high-resourced settings, leading to unreliable outcomes. |
| Approach: | They propose a calibration strategy that takes into account the inherent variance of generalization ability in model components and potential changes during the fine-tuning process. |
| Outcome: | The proposed calibration improves GLUE score by 3.1 points over the previous calibration method. |
SDiaReward: Modeling and Benchmarking Spoken Dialogue Rewards with Modality and Colloquialness (2026.acl-long)
Copied to clipboard
Jingyu Lu, Yuhan Wang, Fan Zhuo, Xize Cheng, Changhao Pan, Xueyi Pu, Yifu Chen, Chenyuhao Wen, Tianle Liang, Zhou Zhao
| Challenge: | SDiaReward is an end-to-end spoken dialogue system that integrates paralinguistic nuances and spontaneous nature of human conversation. |
| Approach: | They propose an end-to-end multi-turn reward model trained on SDiaReward-Dataset . it is a collection of episode-level preference pairs targeting modality and colloquiality gaps . |
| Outcome: | The proposed model outperforms general-purpose audio LLMs in episode-level evaluation. |
TOME: A Two-stage Approach for Model-based Retrieval (2023.acl-long)
Copied to clipboard
| Challenge: | Recent research has focused on model-based retrieval, which discards the index in the traditional retrieval model and memorizes the candidate corpora using model parameters. |
| Approach: | They propose a model-based retrieval approach that discards the index in the traditional retrieval model and memorizes the candidate corpora using model parameters. |
| Outcome: | The proposed approach eliminates the index in the traditional retrieval model and memorizes the candidate corpora using model parameters. |
Skeleton-Guided-Translation: A Benchmarking Framework for Code Repository Translation with Fine-Grained Quality Evaluation (2025.findings-emnlp)
Copied to clipboard
Xing Zhang, Jiaheng Wen, Fangkai Yang, Yu Kang, Pu Zhao, Junhao Wang, Maoquan Wang, Yufan Huang, Shengyu Fu, Elsie Nallipogu, Qingwei Lin, Yingnong Dang, Saravan Rajmohan, Dongmei Zhang
| Challenge: | Existing code translation benchmarks focus on individual functions, overlooking repository-level challenges like intermodule coherence and dependency management. |
| Approach: | They propose a framework for benchmarking Java-to-C# translation at the repository level . it uses a translation framework guided by skeletons and fine-grained quality evaluation . |
| Outcome: | The proposed framework improves Java-to-C# translation quality at the repository level. |
Generating Long and Informative Reviews with Aspect-Aware Coarse-to-Fine Decoding (P19-1)
Copied to clipboard
| Challenge: | Existing methods for review generation lack topical and syntactic characteristics of natural languages. |
| Approach: | They propose a review generation model that uses aspect semantics, syntactic sketch, and context information to generate a sentence and corresponding words. |
| Outcome: | The proposed model can generate long and informative review text for users given a product and her/his rating on it. |
Diffusion-NAT: Self-Prompting Discrete Diffusion for Non-Autoregressive Text Generation (2024.eacl-long)
Copied to clipboard
| Challenge: | Existing non-autoregressive (NAR) text-to-text generation methods are unable to generate coherent and fluent texts due to discrete nature of text. |
| Approach: | They propose to integrate discrete diffusion models (DDM) into NAR text-to-text generation and integrate BART to improve the performance. |
| Outcome: | The proposed method outperforms competing methods and surpasses autoregressive methods on 7 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. |
Debiased Contrastive Learning of Unsupervised Sentence Representations (2022.acl-long)
Copied to clipboard
| Challenge: | Recent studies have shown that contrastive learning improves pre-trained language models to derive high-quality sentence representations. |
| Approach: | They propose a framework to punish false negatives and generate noise-based negatives to guarantee the uniformity of the representation space. |
| Outcome: | The proposed framework improves pre-trained language models while pushing apart irrelevant negatives to guarantee the uniformity of the representation space. |
Evaluating Object Hallucination in Large Vision-Language Models (2023.emnlp-main)
Copied to clipboard
| Challenge: | Large vision-language models (LVLMs) suffer from object hallucinations, i.e., they tend to generate objects inconsistent with the target images in the descriptions. |
| Approach: | They propose to integrate powerful large vision-language models (LVLMs) they propose a polling-based query method to evaluate object hallucination . |
| Outcome: | The proposed model can evaluate object hallucination in a more stable and flexible way. |
DAWN-ICL: Strategic Planning of Problem-solving Trajectories for Zero-Shot In-Context Learning (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing methods to conduct in-context learning without using human-annotated demonstrations are unreliable and lead to error accumulation. |
| Approach: | They propose a method to conduct in-context learning without using human-annotated demonstrations. |
| Outcome: | The proposed method outperforms existing methods using human-annotated demonstrations. |
SHAPE: Unifying Safety, Helpfulness and Pedagogy for Educational LLMs (2026.acl-long)
Copied to clipboard
| Challenge: | Existing educational LLMs are vulnerable to pedagogical jailbreaks where students use answer-inducing prompts to elicit solutions rather than scaffolded instructions. |
| Approach: | They propose a graph-augmented tutoring pipeline that infers prerequisite concepts from queries and identifies mastery gaps. |
| Outcome: | The proposed method improves safety under two pedagogical jailbreak scenarios while maintaining near-ceiling helpfulness under the same evaluation protocol. |
StructGPT: A General Framework for Large Language Model to Reason over Structured Data (2023.emnlp-main)
Copied to clipboard
| Challenge: | Experiments conducted on three types of structured data show that StructGPT greatly improves the performance of LLMs. |
| Approach: | They propose an iterative Reading-then-Reasoning framework to solve question answering tasks based on structured data. |
| Outcome: | The proposed framework improves the reasoning ability of large language models over structured data under the few-shot and zero-shot settings. |
Guiding Neural Entity Alignment with Compatibility (2022.emnlp-main)
Copied to clipboard
| Challenge: | Entity Alignment (EA) aims to find equivalent entities between two Knowledge Graphs (KGs) labelled data is used to learn neural EA models, but this aspect is neglected . |
| Approach: | They propose a framework to integrate compatibility into neural EA models . they aim to find equivalent entities between two Knowledge Graphs (KGs) |
| Outcome: | The proposed framework can achieve comparable effectiveness with supervised training using 20% of labelled data. |
Unveiling the Flaws: Exploring Imperfections in Synthetic Data and Mitigation Strategies for Large Language Models (2024.findings-emnlp)
Copied to clipboard
| Challenge: | acquiring large amounts of high-quality data can be challenging due to data scarcity, privacy concerns, and high costs. |
| Approach: | They propose a method which reverses instruction-following issues caused by uniform format of synthetic data and proposes unlearning techniques to mitigate these flaws. |
| Outcome: | The proposed method reverses instruction-following issues caused by pattern overfitting without compromising performance on benchmarks at relatively low cost. |
Small Pre-trained Language Models Can be Fine-tuned as Large Models via Over-Parameterization (2023.acl-long)
Copied to clipboard
| Challenge: | Large pre-trained language models (PLMs) have shown remarkable performance in various natural language processing tasks, outperforming small PLMs by a large margin. |
| Approach: | They propose to scale up parameters of pre-trained language models only during fine-tuning to benefit from over-parameterization. |
| Outcome: | The proposed approach can significantly boost the fine-tuning performance of small PLMs and even help small PDMs outperform 3 parameterized larger ones. |
Beyond Fully Random Masking: Attention-Guided Denoising and Optimization for Diffusion Language Models (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for full-attention dLLMs rely on random masking strategies that overlook intrinsic token dependencies. |
| Approach: | They propose an attention-guided denoising and optimization framework that aligns training and optimization with attention-derived dependencies. |
| Outcome: | The proposed framework outperforms state-of-the-art methods on mathematical and coding benchmarks. |
A Collaborative Reasoning Framework Powered by Reinforcement Learning and Large Language Models for Complex Questions Answering over Knowledge Graph (2025.coling-main)
Copied to clipboard
| Challenge: | Knowledge Graph Question Answering (KGQA) aims to answer natural language questions by reasoning across multiple triples in knowledge graphs. |
| Approach: | They propose a collaborative reasoning framework powered by RL and LLMs to answer complex questions based on the knowledge graph. |
| Outcome: | The proposed model surpasses state-of-the-art models on four datasets. |
ReFL: Reflective Feedback Learning for Hallucination Detection of Large Language Models (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for detecting hallucinations depend on external knowledge sources, incurring high computational costs and limiting real-time applicability, or extract the model’s internal states, leading to poor generalization. |
| Approach: | They propose a hallucination detection framework that leverages corrective in-context learning to guide LLMs to recognize their own prediction errors and adjust internal representations, critically without updating model weights. |
| Outcome: | The proposed framework outperforms existing methods on two benchmark datasets and achieves state-of-the-art performance. |
KG-Agent: An Efficient Autonomous Agent Framework for Complex Reasoning over Knowledge Graph (2025.acl-long)
Copied to clipboard
| Challenge: | Existing methods to design the interaction strategy between large language models and knowledge graphs (KGs) are not effective for large language model (LLM)s to solve complex tasks due to the large volume and structured format of KG data. |
| Approach: | They propose an LLM-based agent framework that enables small LLMs to actively make decisions over knowledge graphs. |
| Outcome: | The proposed framework outperforms existing methods on in-domain and out-domain datasets using 10K samples. |
Universal Vulnerabilities in Large Language Models: Backdoor Attacks for In-context Learning (2024.emnlp-main)
Copied to clipboard
| Challenge: | In-context learning has shown high efficacy in several NLP tasks, especially in few-shot settings. |
| Approach: | They propose a backdoor attack method that poisons demonstration examples and poisons the demonstration context, preserving the model's generality. |
| Outcome: | The proposed method can make models behave in alignment with predefined intentions without fine-tuning the model. |
GenPilot: A Multi-Agent System for Test-Time Prompt Optimization in Image Generation (2025.findings-emnlp)
Copied to clipboard
Wen Ye, Zhaocheng Liu, Gui Yuwei, Tingyu Yuan, Yunyue Su, Bowen Fang, Chaoyang Zhao, Qiang Liu, Liang Wang
| Challenge: | Existing methods for text-to-image synthesis lack systematic error analysis and refinement strategies, resulting in limited reliability and effectiveness. |
| Approach: | They propose a plug-and-play multi-agent system called GenPilot that integrates error analysis, clustering-based adaptive exploration, fine-grained verification and a memory module for iterative optimization. |
| Outcome: | The proposed method improves text consistency and structural coherence on images with a plug-and-play system. |
Enabling Lightweight Fine-tuning for Pre-trained Language Model Compression based on Matrix Product Operators (2021.acl-long)
Copied to clipboard
| Challenge: | Existing methods for pre-trained language models (PLMs) use parameter reduction techniques. |
| Approach: | They propose a pre-trained language model compression approach based on the matrix product operator from quantum many-body physics. |
| Outcome: | The proposed approach can decompose an original matrix into central tensors and auxiliary tenses . it can be applied to the original or compressed PLMs in a general way, with a lighter network . |
Video-Text Prompting for Weakly Supervised Spatio-Temporal Video Grounding (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods extract each candidate tube feature independently by cropping objects from video frame feature, discarding all contextual information such as position change and inter-entity relationship. |
| Approach: | They propose to use video-text prompts to construct candidate feature instead of cropping tube region from feature map . they also propose negative contrastive samples whose candidate object is erased instead of being highlighted . |
| Outcome: | The proposed methods surpass existing weakly-supervised methods by a great margin . they draw visual markers over objects tubes as video prompts . |
A Neural Citation Count Prediction Model based on Peer Review Text (D19-1)
Copied to clipboard
| Challenge: | Recent studies have focused on extracting or mining useful features from the paper itself or the associated authors. |
| Approach: | They propose to utilize peer review data for the CCP task with a neural prediction model to learn a comprehensive semantic representation for peer review text. |
| Outcome: | The proposed model improves on the peer review data and hand-crafted features. |
VENUS: A VLLM-driven Video Content Discovery System for Real Application Scenarios (2025.emnlp-industry)
Copied to clipboard
Minyi Zhao, Yi Liu, Jianfeng Wen, Boshen Zhang, Hailang Chang, Zhiheng Ouyang, Jie Wang, Wensong He, Shuigeng Zhou
| Challenge: | Video Content Discovery (VCD) is to identify specific videos defined by a pre-specified text policy. |
| Approach: | They propose a Vision-Language Large Model-driven video content discovery system called VENUS to solve these problems. |
| Outcome: | The proposed system generates high-quality, VCD-specific data for model training and extends it to support it better. |
MVP: Multi-task Supervised Pre-training for Natural Language Generation (2023.findings-acl)
Copied to clipboard
| Challenge: | Pre-trained language models (PLMs) have achieved remarkable success in natural language generation tasks. |
| Approach: | They propose to use a large-scale natural language generation corpus to pre-train a text generation model MVP in a supervised manner. |
| Outcome: | The proposed model outperforms BART and Flan-T5 on 13 out of 17 datasets and outperformed BART by 9.3% and FlaN-T5. |
MENTOR: Efficient Autoregressive Image Generation with Balanced Multimodal Control (2026.findings-acl)
Copied to clipboard
| Challenge: | Recent text-to-image models achieve impressive visual quality but still face challenges in precise controllability, balancing multimodal inputs, and high training cost for multimodal image generation. |
| Approach: | They propose an autoregressive framework with a two-stage training paradigm for controllable multimodal image generation. |
| Outcome: | Extensive experiments on DreamBench++ and DreamBech show that the proposed framework achieves a strong balance between textual and visual guidance for controllable image generation. |
AuriSRec: Adversarial User Intention Learning in Sequential Recommendation (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing work focuses on capturing user implicit preferences from historical interactions and matching them with the next behavior, instead of predicting user explicit intentions. |
| Approach: | They propose an adversarial user intention learning approach for sequential recommendaiton . the approach explicitly predicts user current intentions by taking historical reviews as inputs . |
| Outcome: | The proposed approach explicitly predicts user intentions by inferring their decision-making process as explained in target reviews. |
Parameter-Efficient Mixture-of-Experts Architecture for Pre-trained Language Models (2022.coling-1)
Copied to clipboard
| Challenge: | Recent results show that the mix-of-experts architecture is parameter inefficient . large-scale pre-trained language models can achieve excellent performance in many NLP tasks. |
| Approach: | They propose to build a parameter-efficient mix-of-experts architecture by sharing information across experts. |
| Outcome: | The proposed architecture increases model capacity without increasing computation costs. |
Great~Truths~are ~Always ~Simple: A Rather Simple Knowledge Encoder for Enhancing the Commonsense Reasoning Capacity of Pre-Trained Models (2022.findings-naacl)
Copied to clipboard
| Challenge: | Existing approaches to enhance pre-trained language models (PTMs) with a knowledge-aware graph neural network (GNN) encoder that models a commonsense knowledge graph (CSKG) can't explain how external knowledge resources improve the reasoning capacity of PTMs. |
| Approach: | They propose to use relation features from CSKGs to enhance the reasoning capacity of pre-trained language models (PTMs) by encoding a commonsense knowledge graph (CSKG) |
| Outcome: | The proposed approach reduces the parameters for encoding CSKGs and improves on five benchmarks. |
Virtual Data Augmentation: A Robust and General Framework for Fine-tuning Pre-trained Models (2021.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies have shown that powerful pre-trained language models can be fooled by small perturbations or intentional attacks. |
| Approach: | They propose a framework for fine-tuning PLMs using a masked language model and Gaussian noise to augment semantically relevant examples with sufficient diversity. |
| Outcome: | The proposed framework improves the robustness of pre-trained language models and alleviates performance degradation under adversarial attacks. |
Ruleformer: Context-aware Rule Mining over Knowledge Graph (2022.coling-1)
Copied to clipboard
| Challenge: | Existing work on rule mining focuses on mining rules, but how to select appropriate rules for completion of different triplets has not been discussed. |
| Approach: | They propose to take context information into consideration when selecting suitable rules . they devise a transformer-based rule mining approach, Ruleformer . |
| Outcome: | The proposed model takes context information into consideration, which helps select suitable rules for inference tasks. |
Challenging the Boundaries of Reasoning: An Olympiad-Level Math Benchmark for Large Language Models (2026.acl-long)
Copied to clipboard
| Challenge: | Existing evaluation frameworks for large reasoning models are saturated by a lack of reliable and verifiable benchmarks. |
| Approach: | They propose a rigorously curated, Olympiad-level math benchmark comprising 350 problems, each with parallel English and Chinese versions. |
| Outcome: | The proposed benchmark unifies two evaluation paradigms and offers 150 problems formalized in Lean 4 for rigorous process-level evaluation. |
CRSLab: An Open-Source Toolkit for Building Conversational Recommender System (2021.acl-demo)
Copied to clipboard
Kun Zhou, Xiaolei Wang, Yuanhang Zhou, Chenzhan Shang, Yuan Cheng, Wayne Xin Zhao, Yaliang Li, Ji-Rong Wen
| Challenge: | Existing studies on conversational recommender systems lack a unified and standardized implementation or comparison. |
| Approach: | They propose to use a unified framework and highly-decoupled modules to develop CRSs. |
| Outcome: | The proposed framework collects 6 commonly used human-annotated CRS datasets and implements 19 models that include advanced techniques such as graph neural networks and pre-training models. |
Multi-Domain Neural Machine Translation with Word-Level Domain Context Discrimination (D18-1)
Copied to clipboard
| Challenge: | Experimental results on Chinese-English and English-French multi-domain translation tasks demonstrate the effectiveness of the proposed model. |
| Approach: | They propose to use mixed-domain parallel sentences to construct a unified model that allows translation to switch between different domains. |
| Outcome: | The proposed model distinguishes and exploits word-level domain contexts on Chinese-English and English-French translation tasks. |
Rule-KBQA: Rule-Guided Reasoning for Complex Knowledge Base Question Answering with Large Language Models (2025.coling-main)
Copied to clipboard
| Challenge: | Existing methods for knowledge base question answering lack grammaticality, faithfulness, and controllability due to hallucinations in the reasoning process. |
| Approach: | They propose a framework that employs learned rules to guide the generation of logical forms. |
| Outcome: | The proposed method achieves competitive results on standard KBQA datasets. |
Learning How and What to Memorize: Cognition-Inspired Two-Stage Optimization for Evolving Memory (2026.acl-long)
Copied to clipboard
Derong Xu, Shuochen Liu, Pengfei Luo, Pengyue Jia, Yingyi Zhang, Yi Wen, Yimin Deng, Wenlin Zhang, Enhong Chen, Xiangyu Zhao, Tong Xu
| Challenge: | Existing memory systems rely on static, hand-crafted update rules for personalization, but sparse outcome rewards provide weak supervision, resulting in unstable long-horizon optimization. |
| Approach: | They propose a memory guideline optimization framework that learns how memory should be organized and what information to update. |
| Outcome: | The proposed framework learns how memory should be organized and what information to update. |
Attribution-Based Analysis and Optimization of Modular Agentic Workflows (2026.findings-acl)
Copied to clipboard
Yingxuan Yang, Bo Huang, Siyuan Qi, Chao Feng, Haoyi Hu, Yuxuan Zhu, Jinbo Hu, Haoran Zhao, Ziyi He, Xiao Liu, ZongYu Wang, Muning Wen, Lin Qiu, Xuezhi Cao, Xunliang Cai, Yong Yu, Weinan Zhang
| Challenge: | Large Language Models (LLMs) have driven the rise of agentic workflows . yet, how can we attribute performance gains to individual upgrades and their interactions? |
| Approach: | They propose a game-theoretic framework that models component upgrades as players and evaluates component coalitions to compute Shapley values. |
| Outcome: | The proposed framework provides interaction-aware attribution and recommendation for model allocation under a fixed workflow structure. |
LogRules: Enhancing Log Analysis Capability of Large Language Models through Rules (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing large language models (LLMs) exhibit hallucinations when analyzing logs due to the implicit knowledge and rules in logs that LLMs cannot capture. |
| Approach: | They propose a lightweight log analysis framework that generates and utilizes rules through LLMs. |
| Outcome: | The proposed framework outperforms LLM-based methods in log parsing and anomaly detection tasks and achieves better performance compared to case-based approaches. |
ICLEval: Evaluating In-Context Learning Ability of Large Language Models (2025.coling-main)
Copied to clipboard
| Challenge: | Existing evaluation frameworks focus on language abilities and knowledge, often overlooking the assessment of ICL ability. |
| Approach: | They propose to evaluate the ICL ability of Large Language Models (LLMs) using the ICLEval benchmark. |
| Outcome: | The proposed benchmark demonstrates that ICL ability is universally present in different LLMs and model size is not the sole determinant of ICL efficacy. |
Logic-Thinker: Teaching Large Language Models to Think more Logically. (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Recent Large Reasoning Models (LRMs) have demonstrated the ability to generate long chains of thought (LongCoT) LongCoT still faces challenges such as redundancy and logical incoherence. |
| Approach: | They propose a neural-symbolic reasoning framework that generates chains of thought . they propose Logic-Thinker, which transforms symbolic solvers into chains of thoughts . |
| Outcome: | The proposed framework outperforms models fine-tuned with ThinkerCoT on logic reasoning tasks. |
PAIR: Leveraging Passage-Centric Similarity Relation for Improving Dense Passage Retrieval (2021.findings-acl)
Copied to clipboard
Ruiyang Ren, Shangwen Lv, Yingqi Qu, Jing Liu, Wayne Xin Zhao, QiaoQiao She, Hua Wu, Haifeng Wang, Ji-Rong Wen
| Challenge: | Recent studies only consider query-centric similarity relation when learning the dual-encoder retriever. |
| Approach: | They propose a query-centric and PAssage-centric approach to capture more comprehensive similarity relations for dense passage retrieval. |
| Outcome: | The proposed approach significantly outperforms existing models on both MSMARCO and Natural Questions datasets. |
Visual Prompt Tuning for Few-Shot Text Classification (2022.coling-1)
Copied to clipboard
| Challenge: | Existing work on pretraining models for text classification uses image encoders instead of visual prompts. |
| Approach: | They propose a method to deploy large-scale pre-trained models in the prompt-tuning paradigm in few-shot learning. |
| Outcome: | The proposed method outperforms the most recent prompt-tuning methods on five public text classification datasets. |
ViFT: Towards Visual Instruction-Free Fine-tuning for Large Vision-Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Visual instruction tuning is the predominant technology in eliciting multimodal task-solving capabilities of large vision-language models. |
| Approach: | They propose a visual instruction-free fine-tuning framework for large vision-language models . they require only text-only instructions and image caption data during training . |
| Outcome: | The proposed framework is based on visual instruction tuning, but requires images as input . it can achieve state-of-the-art performance on several downstream benchmarks with less training data. |
Finding the Dominant Winning Ticket in Pre-Trained Language Models (2022.findings-acl)
Copied to clipboard
| Challenge: | Existing studies on pre-trained language models show that they can fine-tune parameters but achieve good downstream performance. |
| Approach: | They find that a dominant winning ticket takes up 0.05% of the parameters and is transferable across different tasks. |
| Outcome: | The proposed model can achieve comparable performance with the full-parameter model, the authors show . the dominant winning ticket takes up 0.05% of the parameters, and the model is transferable across tasks, they show - the authors conclude . |
Embracing Large Language Models in Traffic Flow Forecasting (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to predict future traffic flows capture spatio-temporal dependencies, but they fail to adapt to test-time environmental changes. |
| Approach: | They propose to use large language models to help traffic flow forecasting by capturing spatio-temporal dependencies and using a large language model to select the most likely result. |
| Outcome: | The proposed method is based on large language models (LLMs) and an LLM-based selector. |
MMATH: A Multilingual Benchmark for Mathematical Reasoning (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a benchmark for multilingual complex reasoning spans 374 high-quality math problems across 10 typologically diverse languages. |
| Approach: | They propose a benchmark for multilingual complex reasoning across 10 languages . they show reasoning in English and answering in target languages can enhance performance . |
| Outcome: | The proposed benchmark demonstrates that models with high-quality reasoning can perform in multiple languages. |
Comprehensive Benchmarking of Long-Form Speech Generation in Diverse Scenarios (2026.findings-acl)
Copied to clipboard
Changhao Pan, Rui Yang, Han Wang, Zhuan Zhou, Xuming He, Wenxiang Guo, Ziyue Jiang, Ruiqi Li, Yu Zhang, Chenyuhao Wen, Ke Lei, Xiang Yin, Jingyu Lu, Zhiyuan Zhu, Zhou Zhao
| Challenge: | Existing evaluation benchmarks for long-form speech are limited to limited domains, creating a significant gap with the diverse downstream applications. |
| Approach: | They propose a benchmark that decomposes "long-form speech quality" into specific, disentangled dimensions. |
| Outcome: | The proposed benchmark decomposes “long-form speech quality” into specific, disentangled dimensions. |
Unlocking Data-free Low-bit Quantization with Matrix Decomposition for KV Cache Compression (2024.acl-long)
Copied to clipboard
| Challenge: | Existing methods to compress KV cache compromise precision or require extra data for calibration, limiting their practicality in LLM deployment. |
| Approach: | They propose a low-bit quantization technique based on tensor decomposition to effectively compress KV cache. |
| Outcome: | The proposed method reduces memory footprint and performance by 75% . it is compared with existing methods that compromise precision or require extra data for calibration . |
Bloom-Eval: A Hierarchical Evaluation Benchmark for Automatic Survey Generation Based on Bloom’s Taxonomy (2026.acl-long)
Copied to clipboard
| Challenge: | Existing evaluation methods suffer from cognitive dimensional simplification and methodological unreliability due to the ”LLM-as-a-Judge” approach. |
| Approach: | They propose a six-tiered benchmark that evaluates ASG systems by prioritizing deterministic algorithms and introducing a GRADE approach for abstract abilities. |
| Outcome: | The proposed method provides the ASG field with a systematic, reproducible, and theoretically grounded benchmark to guide future research. |
Spec-o3: A Tool-Augmented Vision-Language Agent for Rare Celestial Object Candidate Vetting via Automated Spectral Inspection (2026.acl-long)
Copied to clipboard
| Challenge: | Spec-o3 is a tool-augmented vision-language agent that performs astronomer-aligned spectral inspection. |
| Approach: | They propose a tool-augmented vision-language agent that performs astronomer-aligned spectral inspection via interleaved multimodal chain-of-thought reasoning. |
| Outcome: | Spec-o3 outperforms traditional visual inspection methods on rare-object inspection tasks. |
A Survey of Large Language Models for Text-Guided Molecular Discovery: From Molecule Generation to Optimization (2026.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) are introducing a paradigm shift in molecular discovery by enabling text-guided interaction with chemical spaces through natural language and symbolic notations. |
| Approach: | They analyze the current LLM learning paradigms to tackle four critical evaluation dimensions that have emerged as critical dimensions in recent studies. |
| Outcome: | The proposed models are able to interact with chemical spaces through natural language and symbolic notations, and have emerging extensions to incorporate multi-modal inputs. |
Bayesian Active Learning with Gaussian Processes Guided by LLM Relevance Scoring for Dense Passage Retrieval (2026.findings-acl)
Copied to clipboard
Junyoung Kim, Anton Korikov, Jiazhou Liang, Justin Cui, Yifan Simon Liu, Qianfeng Wen, Mark Zhao, Scott Sanner
| Challenge: | Existing approaches passively rely on first-stage dense retrievers, which leads to two limitations: failing to retrieve relevant passages in semantically distinct clusters and failing to propagate relevance signals to the broader corpus. |
| Approach: | They propose a framework that propagates sparse LLM relevance signals across the embedding space to guide global exploration. |
| Outcome: | Experiments show that the proposed framework outperforms existing approaches under the same budget on all four datasets. |
Extracting and Combining Abilities For Building Multi-lingual Ability-enhanced Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing work relies on training with multi-lingual ability-related data, which may not be available for low-resource languages. |
| Approach: | They propose a multi-lingual ability-enhanced LLM that extracts language-agnostic ability-related weights from LLMs and combine them across different languages by simple addition and subtraction operations without training. |
| Outcome: | The proposed approach extracts language-agnostic ability-related weights from LLMs and combine them across different languages without training. |
YuLan-Mini: Pushing the Limits of Open Data-efficient Language Model (2025.acl-long)
Copied to clipboard
Hu Yiwen, Huatong Song, Jie Chen, Jia Deng, Jiapeng Wang, Kun Zhou, Yutao Zhu, Jinhao Jiang, Zican Dong, Yang Lu, Xu Miao, Xin Zhao, Ji-Rong Wen
| Challenge: | prevailing pre-training approaches for large language models involve several complexities. |
| Approach: | They propose a low-cost training recipe and a robust optimization approach to mitigate training instability . they also propose synthesis, curriculum, and data selection pipelines to integrate data . |
| Outcome: | The proposed model achieves top-tier performance among models with similar parameter scale . it is comparable to industry-leading models that require significantly more data . |
BAMBOO: A Comprehensive Benchmark for Evaluating Long Text Modeling Capacities of Large Language Models (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing long context models suffer from performance decline when the input text exceeds their length limit. |
| Approach: | They propose a multi-task long context benchmark to evaluate LLMs' long context ability using 10 datasets from 5 different NLP tasks. |
| Outcome: | The proposed model covers 5 domains and core capacities of large language models. |
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. |
Defending Against Weight-Poisoning Backdoor Attacks for Parameter-Efficient Fine-Tuning (2024.findings-naacl)
Copied to clipboard
| Challenge: | Existing methods for parameter-efficient fine-tuning (PEFT) are not effective for weight-poisoning backdoor attacks. |
| Approach: | They propose a parameter-efficient fine-tuning (PEFT) method that updates only a limited set of model parameters and provides a robust defense against weight-poisoning backdoor attacks. |
| Outcome: | The proposed method identifies poisoned samples through confidence and is robust against weight-poisoning backdoor attacks. |
SimANS: Simple Ambiguous Negatives Sampling for Dense Text Retrieval (2022.emnlp-industry)
Copied to clipboard
Kun Zhou, Yeyun Gong, Xiao Liu, Wayne Xin Zhao, Yelong Shen, Anlei Dong, Jingwen Lu, Rangan Majumder, Ji-rong Wen, Nan Duan
| Challenge: | Existing methods for sapping negatives from large document pool suffer from the uninformative or false negative problem. |
| Approach: | They propose a method to sample negatives from a large document pool using a new sampling probability distribution. |
| Outcome: | The proposed method can be used to sample more ambiguous negatives on four public and one industry datasets. |
Zero-shot Visual Question Answering with Language Model Feedback (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for knowledge-based visual question answering are based on pre-trained language models. |
| Approach: | They propose a language model guided captioning approach that leverages a pre-trained language model to generate captions for an image to help answer a visual question. |
| Outcome: | The proposed method outperforms several competing methods on the knowledge-based VQA task and achieves comparable results to a fine-tuned VLP model. |
Web Sitemap Knowledge Can Enhance Autonomous Browsing (2026.findings-acl)
Copied to clipboard
Yuyao Zhang, Hongyu Lu, Jiajie Jin, Hongjin Qian, Shiyu Li, Zhao Yang, Yutao Zhu, Ji-Rong Wen, Zhicheng Dou
| Challenge: | Existing web agents suffer from limited robustness, efficiency and task success due to lack of structural understanding of websites and lack of browsing priors in pre-trained models. |
| Approach: | They propose an agent-oriented sitemap protocol that integrates structured website knowledge into web agents. |
| Outcome: | The proposed agent-oriented sitemap improves robustness, efficiency and effectiveness without extra training. |
The Dawn After the Dark: An Empirical Study on Factuality Hallucination in Large Language Models (2024.acl-long)
Copied to clipboard
| Challenge: | a growing number of researchers are studying the hallucination issue in large language models. |
| Approach: | They propose a hallucination detection benchmark and a method to detect hallucines in LLMs. |
| Outcome: | The proposed method detects hallucinations and mitigates them using different training stages. |
Sticker-TTS: Learn to Utilize Historical Experience with a Sticker-driven Test-Time Scaling Framework (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large reasoning models have exhibited strong performance on complex reasoning tasks, but current test-time scaling methods rely on redundant sampling and ignore historical experience utilization. |
| Approach: | They propose a test-time scaling framework that coordinates three collaborative LRMs to iteratively explore and refine solutions guided by historical attempts. |
| Outcome: | The proposed framework surpasses strong baselines on three mathematical reasoning benchmarks, including AIME-24, AIME-25, and OlymMATH. |
Unilaw-R1: A Large Language Model for Legal Reasoning with Reinforcement Learning and Iterative Inference (2025.emnlp-main)
Copied to clipboard
| Challenge: | Reasoning-focused large language models (LLMs) are rapidly evolving across various domains, yet their capabilities in handling complex legal problems remain underexplored. |
| Approach: | They propose a large language model tailored for legal reasoning with a 7-billion parameter scale and a two-stage training strategy combining Supervised Fine-Tuning and Reinforcement Learning. |
| Outcome: | The proposed model outperforms all models of similar scale on authoritative benchmarks and outperformed Qwen-2.5-7B-Instruct (46.6%) by an average margin of 6.6%. |
Smart-Searcher: Incentivizing the Dynamic Knowledge Acquisition of LLMs via Reinforcement Learning (2025.findings-emnlp)
Copied to clipboard
Huatong Song, Jinhao Jiang, Wenqing Tian, Zhipeng Chen, Yuhuan Wu, Jiahao Zhao, Yingqian Min, Xin Zhao, Lei Fang, Ji-Rong Wen
| Challenge: | Large Language Models (LLMs) are powerful but prone to hallucinations due to static knowledge. Retrieval-augmented generation (RAG) helps by injecting external information, but current methods are costly, generalize poorly, or ignore the model’s internal knowledge. |
| Approach: | They propose a framework to train large language models to leverage both internal and external knowledge sources. |
| Outcome: | The proposed framework outperforms existing methods and achieves efficient retrieval-augmented reasoning. |
Gamma-Guard: Lightweight Residual Adapters for Robust Guardrails in Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) are widely deployed as zero-shot evaluators for answer grading, content moderation, and document ranking. |
| Approach: | They propose a system that trains LLMs with adapters to denoise embeddings and refocus attention. |
| Outcome: | The proposed model lifts adversarial accuracy from 5% to 95% a 90 percentage-point gain while reducing clean-data accuracy by just 8 percentage points. |
QiMeng-Attention: SOTA Attention Operator is generated by SOTA Attention Algorithm (2025.findings-acl)
Copied to clipboard
Qirui Zhou, Shaohui Peng, Weiqiang Xiong, Haixin Chen, Yuanbo Wen, Haochen Li, Ling Li, Qi Guo, Yongwei Zhao, Ke Gao, Ruizhi Chen, Yanjun Wu, Zhao Chen, Yunji Chen
| Challenge: | Existing LLMs cannot comprehend the complex data flow and computation process of the attention operator and utilize low-level primitive to exploit GPU performance. |
| Approach: | They propose an LLM-friendly Thinking Language (LLM-TL) that can decouple the generation of high-level optimization logic and low-level implementation on GPU and enhance LLMs’ understanding of attention operator. |
| Outcome: | The proposed method outshines existing LLMs on A100, RTX8000, and T4 GPUs, achieving a speed-up of up to 35.16. |
Towards Effective and Efficient Continual Pre-training of Large Language Models (2025.acl-long)
Copied to clipboard
Jie Chen, Zhipeng Chen, Jiapeng Wang, Kun Zhou, Yutao Zhu, Jinhao Jiang, Yingqian Min, Xin Zhao, Zhicheng Dou, Jiaxin Mao, Yankai Lin, Ruihua Song, Jun Xu, Xu Chen, Rui Yan, Zhewei Wei, Di Hu, Wenbing Huang, Ji-Rong Wen
| Challenge: | Continual pre-training (CPT) has been an important approach for adapting language models to specific domains or tasks. |
| Approach: | They propose a Continual pre-training method that can greatly improve Chinese language ability and scientific reasoning ability of LLMs. |
| Outcome: | The proposed method can greatly improve Chinese language ability and scientific reasoning ability of LLMs. |
Investigating the Factual Knowledge Boundary of Large Language Models with Retrieval Augmentation (2025.coling-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) have shown impressive prowess in solving a wide range of tasks with world knowledge, but it remains unclear how well they perceive their factual knowledge boundaries. |
| Approach: | They propose to use a retrieval augmentation approach to enhance LLMs' awareness of factual knowledge boundaries to analyze factual and factual information in open-domain question answering (QA) |
| Outcome: | The proposed method improves LLMs’ QA and judgemental capabilities by integrating supporting documents with the questions. |
Do Emergent Abilities Exist in Quantized Large Language Models: An Empirical Study (2024.lrec-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) require significant computational resources for deployment and use. |
| Approach: | They propose to use low-bit quantization methods to reduce memory footprint and increase inference rate to improve performance of Large Language Models. |
| Outcome: | The proposed methods can reduce the memory footprint and increase the inference rate of LLMs. |
RocketQAv2: A Joint Training Method for Dense Passage Retrieval and Passage Re-ranking (2021.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies show that passage retrieval and passage reranking are important for achieving mutual improvement. |
| Approach: | They propose a unified listwise training approach for passage retrieval and passage reranking that incorporates a retrieval procedure and a hybrid data augmentation strategy. |
| Outcome: | The proposed approach improves on both MSMARCO and Natural Questions datasets. |
Enhancing Scalability of Pre-trained Language Models via Efficient Parameter Sharing (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing approaches to scale pre-trained language models to a deeper model depth share all parameters or use extra blocks. |
| Approach: | They propose a parameter-efficient approach to scaling pre-trained language models to a deeper model depth using matrix product operator. |
| Outcome: | The proposed model scales pre-trained language models to a deeper model depth by 4x and achieves 0.1 points higher than BERT-large for GLUE score. |
Experience-Driven Reflective Co-Evolution of Prompts and Heuristics for Autonomous Algorithm Design (2026.findings-acl)
Copied to clipboard
| Challenge: | Combinatorial optimization has long been dominated by manually engineered heuristics, which require substantial expert intuition and implementation overhead. |
| Approach: | They propose a framework that couples an island migration model with elite selection to maintain population diversity. |
| Outcome: | The proposed framework achieves superior accuracy on the Traveling Salesman and Bin Packing Problems. |
Continual Pre-training of Language Models for Math Problem Understanding with Syntax-Aware Memory Network (2022.acl-long)
Copied to clipboard
| Challenge: | a fundamental challenge in modeling math problems is how to fuse semantics of textual description and formulas. |
| Approach: | They propose a method to continually pre-train language models for improving understanding of math problems with syntax-aware memory networks. |
| Outcome: | The proposed approach outperforms competitive baselines on four math tasks. |
MarkLLM: An Open-Source Toolkit for LLM Watermarking (2024.emnlp-demo)
Copied to clipboard
Leyi Pan, Aiwei Liu, Zhiwei He, Zitian Gao, Xuandong Zhao, Yijian Lu, Binglin Zhou, Shuliang Liu, Xuming Hu, Lijie Wen, Irwin King, Philip Yu
| Challenge: | Large Language Models (LLMs) embed imperceptible yet algorithmically detectable signals in outputs to identify LLM-generated text. |
| Approach: | They propose to develop an open-source toolkit for LLM watermarking that embeds imperceptible yet algorithmically detectable signals in model outputs to identify LLM-generated text. |
| Outcome: | MarkLLM provides a unified framework for implementing LLM watermarking algorithms, while providing user-friendly interfaces to ensure ease of access. |
ActiveEA: Active Learning for Neural Entity Alignment (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to combining knowledge Graphs (KGs) are incomplete but complementary to each other. |
| Approach: | They propose a novel Active Learning framework for neural EA that creates highly informative seed alignments to obtain more effective models with less annotation cost. |
| Outcome: | The proposed framework significantly improves sampling quality with good generality across different datasets, EA models and amount of bachelors. |
ControlSpeech: Towards Simultaneous and Independent Zero-shot Speaker Cloning and Zero-shot Language Style Control (2025.acl-long)
Copied to clipboard
Shengpeng Ji, Qian Chen, Wen Wang, Jialong Zuo, Minghui Fang, Ziyue Jiang, Hai Huang, Zehan Wang, Xize Cheng, Siqi Zheng, Zhou Zhao
| Challenge: | Prior zero-shot TTS models only mimic the speaker’s voice without further control and adjustment capabilities while prior controllable TTS systems cannot perform speaker-specific voice generation. |
| Approach: | They propose a style control module that captures codec representations corresponding to timbre, content, and style in a discrete decoupling codec space. |
| Outcome: | The proposed system can fully clone the speaker's voice and perform speech-specific adjustment and control functions. |
Learning to Transfer Prompts for Text Generation (2022.naacl-main)
Copied to clipboard
| Challenge: | Pretrained language models (PLMs) have made remarkable progress in text generation tasks via fine-tuning. |
| Approach: | They propose a prompt-based method that learns source prompts and transfers them as target prompts to perform target generation tasks. |
| Outcome: | The proposed method can be used to perform text generation tasks in a transferable setting. |
Towards Topic-Guided Conversational Recommender System (2020.coling-main)
Copied to clipboard
| Challenge: | Existing CRS datasets focus on immediate requests from users, while lack proactive guidance to the recommendation scenario. |
| Approach: | They propose a topic-guided conversational recommendation dataset . it incorporates topic threads to enforce natural semantic transitions towards the recommendation scenario . |
| Outcome: | The proposed approach is more reasonable and controllable than previous approaches. |
REAR: A Relevance-Aware Retrieval-Augmented Framework for Open-Domain Question Answering (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to extend knowledge scope of large language models (LLMs) lack internal parametric knowledge, resulting in misusing external knowledge. |
| Approach: | They propose a retrieval-augmented approach that provides LLMs with potentially relevant documents through a module. |
| Outcome: | The proposed approach outperforms existing methods on four open-domain QA tasks. |
Domain Adaptation for Person-Job Fit with Transferable Deep Global Match Network (D19-1)
Copied to clipboard
| Challenge: | Existing methods to solve person-job fit in single-domain setting are limited by labeled data. |
| Approach: | They propose a deep global match network for capturing the global semantic interactions between two sentences from a job posting and a candidate resume respectively. |
| Outcome: | The proposed model is effective when there is not enough labeled data. |
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. |
Rethinking the Evaluation for Conversational Recommendation in the Era of Large Language Models (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing evaluation protocols for large language models (LLMs) are inadequate for conversational recommender systems. |
| Approach: | They propose an evaluation approach based on LLMs that harnesses LLM-based user simulators to evaluate ChatGPT's performance. |
| Outcome: | The proposed evaluation approach can simulate various system-user interaction scenarios. |
ChatCoT: Tool-Augmented Chain-of-Thought Reasoning on Chat-based Large Language Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models (LLMs) have excellent performance in evaluation benchmarks, but struggle in complex reasoning tasks. |
| Approach: | They propose a tool-augmented chain-of-thought reasoning framework for chat-based LLMs . they model chain- of-thoughting reasoning as multi-turn conversations to utilize tools . |
| Outcome: | The proposed framework can outperform state-of-the-art models on complex reasoning tasks. |
Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models (2024.acl-long)
Copied to clipboard
Tianyi Tang, Wenyang Luo, Haoyang Huang, Dongdong Zhang, Xiaolei Wang, Xin Zhao, Furu Wei, Ji-Rong Wen
| Challenge: | Despite the impressive multilingual capabilities demonstrated by LLMs, the understanding of how these abilities develop and function remains nascent. |
| Approach: | They propose a novel detection method to pinpoint language-specific neurons within LLMs by selectively activating or deactivating these neurons. |
| Outcome: | The proposed method can “steer” the output language of LLMs by selectively activating or deactivating language-specific neurons. |
Prompt as Triggers for Backdoor Attack: Examining the Vulnerability in Language Models (2023.emnlp-main)
Copied to clipboard
| Challenge: | ProAttack is a novel and efficient method for performing clean-label backdoor attacks based on the prompt, which uses the prompt itself as a trigger. |
| Approach: | They propose a method for performing clean-label backdoor attacks based on the prompt, which uses the prompt itself as a trigger. |
| Outcome: | The proposed method achieves state-of-the-art performance on several NLP tasks, particularly in few-shot settings. |
BalanceSFT: Improving LLM Function Calling with Balanced Training Signals and Data Hardness (2026.findings-acl)
Copied to clipboard
Bingguang Hao, Zengzhuang Xu, Maolin Wang, Yuntao Wen, Yicheng Chen, Cunyin Peng, Long Chen, Xiangyu Zhao, Jinjie Gu, Chenyi Zhuang, Ji Zhang
| Challenge: | Currently, Supervised Fine-Tuning (SFT) is the prevailing method for equipping Large Language Models (LLMs) with function calling capabilities, but its effectiveness is often compromised by two challenges: 1) lengthy Chain-of-Thought (CoT) reasoning tokens dominate training signals over concise function calls in the learning objective; 2) scarcity of hard training examples. |
| Approach: | They propose a framework that uses a self-adjusted signal balancing loss and a hard data re-sampling strategy to selectively generate new, high-quality complex data guided by model errors. |
| Outcome: | The proposed framework surpasses state-of-the-art models like GPT-5 in function calling performance. |
A Thorough Examination on Zero-shot Dense Retrieval (2023.findings-emnlp)
Copied to clipboard
Ruiyang Ren, Yingqi Qu, Jing Liu, Xin Zhao, Qifei Wu, Yuchen Ding, Hua Wu, Haifeng Wang, Ji-Rong Wen
| Challenge: | Recent advances in dense retrieval (DR) models have been shown to be not as competitive as traditional sparse retrieval models in a zero-shot retrieval setting. |
| Approach: | They propose to examine the zero-shot capability of DR models by analyzing key factors related to source training set and potential bias from target dataset. |
| Outcome: | The proposed model is not as competitive as sparse retrieval models in a zero-shot retrieval setting. |
DopplerBAS: Binaural Audio Synthesis Addressing Doppler Effect (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for binaural audio synthesis are limited in phase estimation, which is crucial for spatial hearing. |
| Approach: | They propose a method to explicitly address the Doppler effect of the moving speaker . it calculates the radial relative velocity of the speaker in spherical coordinates . |
| Outcome: | The proposed method improves the representative WarpNet and BinauralGrad backbones in phase error metric and reaches a new state of the art (SOTA) it is compared with the current method which is limited in phase estimation . |
HaluEval: A Large-Scale Hallucination Evaluation Benchmark for Large Language Models (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing large language models (LLMs) are prone to generate hallucinations . a recent study shows that LLMs are able to generate content that conflicts with the source or cannot be verified by factual knowledge. |
| Approach: | They propose a framework to evaluate the performance of large language models (LLMs) they propose to use a sample of generated and human-annotated hallucinated samples to evaluate their performance . |
| Outcome: | The proposed framework generates and annotates hallucinated samples from ChatGPT . the results show that existing LLMs face great challenges in recognizing hallucines . |
Interactive-KBQA: Multi-Turn Interactions for Knowledge Base Question Answering with Large Language Models (2024.acl-long)
Copied to clipboard
| Challenge: | Knowledge base question answering (KBQA) is a challenging task, particularly in parsing intricate questions into executable logical forms. |
| Approach: | They propose a framework to generate logical forms through direct interaction with knowledge bases (KBs) by annotating a dataset with step-wise reasoning processes. |
| Outcome: | The proposed framework achieves competitive results on the WebQuestionsSP, ComplexWebQuestIONS, KQA Pro, and MetaQA datasets with a minimal number of examples (shots). Importantly, the proposed model supports manual intervention, allowing for the iterative refinement of LLM outputs. |
TextBox 2.0: A Text Generation Library with Pre-trained Language Models (2022.emnlp-demos)
Copied to clipboard
Tianyi Tang, Junyi Li, Zhipeng Chen, Yiwen Hu, Zhuohao Yu, Wenxun Dai, Wayne Xin Zhao, Jian-yun Nie, Ji-rong Wen
| Challenge: | TextBox 2.0 focuses on the use of pre-trained language models (PLMs) to generate text. |
| Approach: | They propose a library that integrates pre-trained language models into 13 common text generation tasks and 83 datasets. |
| Outcome: | The proposed library covers 13 common text generation tasks and their corresponding datasets and incorporates 45 PLMs covering general, translation, Chinese, dialogue, controllable, distilled, prompting, and lightweight PLM. |
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. |
BASES: Large-scale Web Search User Simulation with Large Language Model based Agents (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing research on web search rely on real-user experiments, which can be costly to scale up. |
| Approach: | They propose a user simulation framework with LLM-based agents that can generate unique user profiles at scale. |
| Outcome: | The proposed framework can generate unique user profiles at scale, leading to diverse search behaviors. |
Point, Disambiguate and Copy: Incorporating Bilingual Dictionaries for Neural Machine Translation (2021.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to incorporate bilingual dictionaries into Neural Machine Translation (NMT) models have been criticized for lack of integration of bilingual lexical information into the neural architecture. |
| Approach: | They propose a neural architecture to incorporate bilingual dictionaries into Neural Machine Translation models by introducing three new components: Pointer, Disambiguator, and Copier. |
| Outcome: | The proposed method achieves the following merits inherently compared with previous efforts: (1) Pointer leverages the semantic information from bilingual dictionaries, for the first time, to better locate source words whose translation in dictionary can potentially be used; (2) Disambiguator synthesizes contextual information from source view and target view, both of which contribute to distinguishing translation of a specific source word from multiple candidates in dicaries; (3) Copier systematically connects Pointer and Disambiguators based on a hierarchical |
The Web Can Be Your Oyster for Improving Language Models (2023.findings-acl)
Copied to clipboard
| Challenge: | Pretrained language models encode a large amount of knowledge, but knowledge is frozen at the time of training, and the models become static and limited by training data. |
| Approach: | They propose an adaptive search engine assisted learning method that can self-evaluate the confidence level of PLM’s predictions and adaptively determine when to refer to the web for more data. |
| Outcome: | The proposed model outperforms retrieval-augmented methods on 16 knowledge-intensive tasks on a wide range of knowledge-related tasks. |
ReasoningLM: Enabling Structural Subgraph Reasoning in Pre-trained Language Models for Question Answering over Knowledge Graph (2023.emnlp-main)
Copied to clipboard
| Challenge: | Question Answering over Knowledge Graph (KGQA) aims to find answer entities for natural language questions based on knowledge graphs. |
| Approach: | They propose a subgraph-aware self-attention mechanism to imitate the graph neural network (GNN) based module to perform multi-hop reasoning on KG. |
| Outcome: | The proposed method surpasses state-of-the-art models by a large margin even with fewer updated parameters and less training data. |
WavAlign: Enhancing Intelligence and Expressiveness in Spoken Dialogue Models via Adaptive Hybrid Post-Training (2026.findings-acl)
Copied to clipboard
Yifu Chen, Shengpeng Ji, Qian Chen, Tianle Liang, Yangzhuo Li, Ziqing Wang, Wen Wang, Jingyu Lu, Haoxiao Wang, Xueyi Pu, Fan Zhuo, Zhou Zhao
| Challenge: | End-to-end spoken dialogue models have higher potential ceiling in expressiveness and perceptual ability than cascaded systems. |
| Approach: | They propose a modality-aware adaptive post-training recipe that constrains preference updates to the semantic channel and improves acoustic behavior via explicit anchoring. |
| Outcome: | The proposed model improves speech quality and expressiveness across spoken dialogue benchmarks and architectures. |
What Makes for Good Visual Instructions? Synthesizing Complex Visual Reasoning Instructions for Visual Instruction Tuning (2025.coling-main)
Copied to clipboard
Yifan Du, Hangyu Guo, Kun Zhou, Wayne Xin Zhao, Jinpeng Wang, Chuyuan Wang, Mingchen Cai, Ruihua Song, Ji-Rong Wen
| Challenge: | Experimental results show that visual instruction tuning improves performance of Multi-modal Large Language Models (MLLMs) to extend the application scope of Large Language Modells, a surge of work augments LLMs with vision encoders to endow the ability of multi-modal cognition and reasoning. |
| Approach: | They propose a systematic approach to create high-quality visual reasoning instructions using a synthesize-complicate-reformulate paradigm. |
| Outcome: | The proposed method improves performance of MLLMs by 27.86% and 27.60% on MME-Perception and MME Cognition. |
SimpleDeepSearcher: Deep Information Seeking via Web-Powered Reasoning Trajectory Synthesis (2025.findings-emnlp)
Copied to clipboard
Shuang Sun, Huatong Song, Yuhao Wang, Ruiyang Ren, Jinhao Jiang, Junjie Zhang, Fei Bai, Jia Deng, Xin Zhao, Zheng Liu, Lei Fang, Zhongyuan Wang, Ji-Rong Wen
| Challenge: | Existing approaches to deep search training lack high-quality training trajectories, prohibitive computational costs and lack of high-fidelity training data. |
| Approach: | They propose a framework that synthesizes high-quality training data by simulating real user interactions in live web search environments. |
| Outcome: | The proposed framework synthesizes high-quality training data by simulating user interactions in live web search environments. |
ChainLM: Empowering Large Language Models with Improved Chain-of-Thought Prompting (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing CoT synthesis approaches focus on simpler reasoning tasks and result in inconsistent CoT prompts. |
| Approach: | They propose a framework for automatic generation of superior CoT prompts based on three major evolution strategies . they propose 'step-level debating' method where multiple debaters discuss each reasoning step to arrive at the correct answer. |
| Outcome: | The proposed framework can generate superior CoT prompts from a CoT dataset. |
Visually-augmented pretrained language models for NLP tasks without images (2023.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to improve pre-trained language models lack visual commonsense and semantics. |
| Approach: | They propose a visual-augmented approach to fine-tune pre-trained language models by using retrieved or generated images instead of relying on explicit images. |
| Outcome: | The proposed approach outperforms baselines on ten tasks and consistently outperformed other approaches. |
MA-DPR: Manifold-aware Distance Metrics for Dense Passage Retrieval (2025.emnlp-main)
Copied to clipboard
| Challenge: | Empirical evidence suggests that manifold-aware distance allows DPR to leverage context from related neighboring passages. |
| Approach: | They propose a "manifold-aware" distance metric that measures query–passage distance . they propose to model the intrinsic manifold structure of passages using a nearest-neighbor graph . |
| Outcome: | Empirical evidence suggests MA-DPR outperforms Euclidean and cosine distances on OOD . it can be applied to a wide range of dense embedding and retrieval tasks . |
TreeMAN: Tree-enhanced Multimodal Attention Network for ICD Coding (2022.coling-1)
Copied to clipboard
| Challenge: | Existing methods to automatically assign ICD codes ignore crucial information contained in structured medical data, which is hard to be captured from the noisy clinical notes. |
| Approach: | They propose to use a Tree-enhanced multimodal attention network to fuse tabular features and textual features into multimodal representations by enhancing the text representations with tree-based features. |
| Outcome: | The proposed method outperforms state-of-the-art methods on two MIMIC datasets. |
Improving Large Language Models via Fine-grained Reinforcement Learning with Minimum Editing Constraint (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing reinforcement learning methods do not provide fine-grained supervision for complex reasoning tasks. |
| Approach: | They propose a reinforcement learning method that incorporates a generative model as the reward model and a token-level supervision model for RL training. |
| Outcome: | Experiments on 8 tasks show the proposed method is effective . |
Not Everything is All You Need: Toward Low-Redundant Optimization for Large Language Model Alignment (2024.emnlp-main)
Copied to clipboard
| Challenge: | Experimental results show that large language models are struggling to align with human preference in complex tasks and scenarios. |
| Approach: | They propose a low-redundant alignment method that selects the top-10% most updated parameters in LLMs for alignment training. |
| Outcome: | The proposed method improves on 10 datasets and shows that it is redundant . it can be used to train LLMs on QA and ECQA datasets, but it is not feasible to test it on a large dataset. |
Large Language Models are Complex Table Parsers (2023.emnlp-main)
Copied to clipboard
| Challenge: | Extensive experiments and results on Complex Table QA datasets, i.e., the open-domain dataset HiTAB and the aviation domain dataset AIT-QA show that our approach significantly outperforms previous work on both datasets. |
| Approach: | They propose to incorporate Generative Pre-trained Transformer 3.5 to address the specific challenges posed by Complex Table QA by reconstructing tables into tuples and using prompt templates to create dialogues. |
| Outcome: | The proposed approach outperforms previous work on complex table parsing datasets and leads to state-of-the-art (SOTA) performance. |
Think More, Hallucinate Less: Mitigating Hallucinations via Dual Process of Fast and Slow Thinking (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing studies have focused on the issue of hallucination in large language models. |
| Approach: | They propose a framework that allows an explicit slow thinking generation process for mitigating hallucinations during inference. |
| Outcome: | The proposed framework outperforms baseline approaches on English and Chinese datasets. |
SSP: Self-Supervised Post-training for Conversational Search (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to search for information are limited by data scarcity and lack contextual semantic understanding. |
| Approach: | They propose a new post-training paradigm with three self-supervised tasks to initialize the conversational search model to enhance dialogue structure and contextual semantic understanding. |
| Outcome: | The proposed method can initialize the conversational search model to enhance dialogue structure and contextual semantic understanding. |
Small Agent Can Also Rock! Empowering Small Language Models as Hallucination Detector (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies on hallucination detection rely heavily on closed-source LLMs such as GPT-4. |
| Approach: | They propose an LLM-based agent framework called HaluAgent that integrates LLMs, multi-functional toolbox and a memory mechanism for hallucination detection. |
| Outcome: | The proposed framework integrates the LLM, multi-functional toolbox, and can detect hallucinations on Chinese and English datasets. |
Few-shot Knowledge Graph-to-Text Generation with Pretrained Language Models (2021.findings-acl)
Copied to clipboard
| Challenge: | Existing models for KG-to-text generation are based on pretrained language models. |
| Approach: | They propose to automatically generate a text that describes the facts in knowledge graph (KG) they leverage the excellent capacities of pretrained language models (PLMs) in language understanding and generation. |
| Outcome: | The proposed model outperforms all comparison methods on fully-supervised and fewshot settings. |
TextBox: A Unified, Modularized, and Extensible Framework for Text Generation (2021.acl-demo)
Copied to clipboard
Junyi Li, Tianyi Tang, Gaole He, Jinhao Jiang, Xiaoxuan Hu, Puzhao Xie, Zhipeng Chen, Zhuohao Yu, Wayne Xin Zhao, Ji-Rong Wen
| Challenge: | TextBox is an open-source text generation framework that is modularized and extensible. |
| Approach: | They propose to provide a unified, modularized, and extensible text generation framework that implements 21 text generation models on 9 benchmark datasets. |
| Outcome: | The proposed framework implements 21 models on 9 benchmark datasets and is available under the Apache License 2.0 license. |
Multimodal Item Scoring for Natural Language Recommendation via Gaussian Process Regression with LLM Relevance Judgments (2026.findings-acl)
Copied to clipboard
Yifan Simon Liu, Qianfeng Wen, Jiazhou Liang, Mark Zhao, Justin Cui, Anton Korikov, Armin Toroghi, Junyoung Kim, Scott Sanner
| Challenge: | Existing NLRec approaches use Dense Retrieval to compute item relevance scores . DR views the request as the sole relevance label, leading to a weak proxy for query relevance. |
| Approach: | They propose to use Gaussian Process Regression to model item relevance . they propose to combine LLM with a Gauss-based kernel to model multimodal relevance judging . |
| Outcome: | The proposed approach outperforms simpler unimodal kernels and baseline methods by up to 65% on four NLRec datasets and two LLM backbones. |
ElitePLM: An Empirical Study on General Language Ability Evaluation of Pretrained Language Models (2022.naacl-main)
Copied to clipboard
Junyi Li, Tianyi Tang, Zheng Gong, Lixin Yang, Zhuohao Yu, Zhipeng Chen, Jingyuan Wang, Xin Zhao, Ji-Rong Wen
| Challenge: | Recent years have featured a trend towards Transformer based pretrained language models (PLMs) in natural language processing systems. |
| Approach: | They propose to use four evaluation dimensions to evaluate ten widely-used PLMs . they find that pretrained language models are good at different ability tests . |
| Outcome: | The results show that pretrained language models are good at different ability tests and have excellent transferability between tasks. |
ELMER: A Non-Autoregressive Pre-trained Language Model for Efficient and Effective Text Generation (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for text generation use auto-regressive (AR) methods, but inefficient inference is a problem. |
| Approach: | They propose an efficient and effective PLM to explicitly model the token dependency during NAR text generation. |
| Outcome: | The proposed model outperforms existing models on three text generation tasks while achieving 10 times faster inference speedup. |