Papers by Xi Chen
UniCodec: Unified Audio Codec with Single Domain-Adaptive Codebook (2025.acl-long)
Copied to clipboard
Yidi Jiang, Qian Chen, Shengpeng Ji, Yu Xi, Wen Wang, Chong Zhang, Xianghu Yue, ShiLiang Zhang, Haizhou Li
| Challenge: | Existing neural audio codecs are not capable of handling multi-domain audio data . et al., 2023) integrate speech modality with text-based large language models . |
| Approach: | They propose a unified audio codec with a single codebook to support multi-domain audio data . they propose combining a mix-of-experts strategy and a partitioned domain-adaptive codebook method . |
| Outcome: | The proposed codec outperforms existing codecs on acoustic and semantic representation capabilities. |
To Forget or Not? Towards Practical Knowledge Unlearning for Large Language Models (2024.findings-emnlp)
Copied to clipboard
Bozhong Tian, Xiaozhuan Liang, Siyuan Cheng, Qingbin Liu, Mengru Wang, Dianbo Sui, Xi Chen, Huajun Chen, Ningyu Zhang
| Challenge: | Existing unlearning paradigms are mired in vague forgetting boundaries, erasing knowledge indiscriminately. |
| Approach: | They propose a benchmark to evaluate if unlearning erases essential knowledge . they propose 'knowUnDo' which uses copyrighted content and privacy domains . |
| Outcome: | The proposed method is superior to existing methods in both precise knowledge unlearning and general knowledge retaining of LLMs. |
Better Process Supervision with Bi-directional Rewarding Signals (2025.findings-acl)
Copied to clipboard
Wenxiang Chen, Wei He, Zhiheng Xi, Honglin Guo, Boyang Hong, Jiazheng Zhang, Nijun Li, Tao Gui, Yun Li, Qi Zhang, Xuanjing Huang
| Challenge: | Existing processes that reward for each step are one-directional and lack a mechanism to model the distance to the final target. |
| Approach: | They propose a process supervision model that evaluates the correctness of previous steps and the probability of future success. |
| Outcome: | The proposed model outperforms existing supervision models like ORM and PRM on reasoning tasks and improves solution re-design. |
Exploring the Impact of Personality Traits on LLM Toxicity and Bias (2025.emnlp-main)
Copied to clipboard
| Challenge: | anthropomorphic LLMs are being developed to serve diversified roles, but content safety concerns remain regarding their toxicity and toxicity. |
| Approach: | They propose to assign personality traits to large language models (LLMs) to reduce toxic language and social biases in their outputs by using the widely accepted HEXACO personality framework developed in social psychology. |
| Outcome: | The proposed model is able to perform on three toxic and bias benchmarks and shows that assigning personality traits reduces bias and toxicity similar to humans’ correlations between personality traits and toxic behaviors. |
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. |
AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments (2026.acl-long)
Copied to clipboard
Zhiheng Xi, Dingwen Yang, Jiaqi Liu, Jixuan Huang, Honglin Guo, Baodai Huang, Tinggang Chen, Qi Zhang, Zhonghang Lu, Chenyu Liu, Jiajun Sun, Jiazheng Zhang, Dingwei Zhu, Xin Guo, Junzhe Wang, Zhihao Zhang, Yuming Yang, Junjie Ye, Minghe Gao, Dongrui Liu, Jiaming Ji, Guohao Li, Tao Gui, Qi Zhang, Xuanjing Huang
| Challenge: | Existing benchmarks evaluate agents in simplified, idealized settings, relying on pre-packaged tool interfaces, overlooking critical steps, and assume inputs are clean and fully specified. |
| Approach: | They propose a framework that evaluates language agents in simplified, idealized settings . they show that even SOTA systems like Gemini and GPT-5 struggle on AgentGym2 . |
| Outcome: | Experiments on 15 proprietary and open-source models show that even SOTA systems like Gemini and GPT-5 struggle on AgentGym2 . |
Chain-of-Scrutiny: Detecting Backdoor Attacks for Large Language Models (2025.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated impressive capabilities across various domains, but are vulnerable to backdoor attacks. |
| Approach: | They propose a chain-of-scrutiny approach which leverages LLMs’ unique reasoning abilities to mitigate backdoor attacks. |
| Outcome: | The proposed model is well-suited for the popular API-only LLM deployments, enabling detection at minimal cost and with little data. |
MaXM: Towards Multilingual Visual Question Answering (2023.findings-emnlp)
Copied to clipboard
Soravit Changpinyo, Linting Xue, Michal Yarom, Ashish Thapliyal, Idan Szpektor, Julien Amelot, Xi Chen, Radu Soricut
| Challenge: | Visual Question Answering (VQA) has been studied in the English language, but in other languages it would require a considerable amount of resources. |
| Approach: | They propose scalable solutions to multilingual visual question answering using an English language framework and an annotation protocol. |
| Outcome: | The proposed framework reduces human annotation efforts and creates a test-only VQA benchmark in 7 languages. |
Learning from Contrasts: Synthesizing Reasoning Paths from Diverse Search Trajectories (2026.acl-long)
Copied to clipboard
| Challenge: | MCTS methods retain only the single highest-reward trajectory, discarding comparative signals present in the many explored paths. |
| Approach: | They propose a framework that transforms supervision extraction into a synthesis procedure. |
| Outcome: | The proposed framework matches or exceeds baselines on 60K CRPS-synthesized examples on out-of-domain benchmarks. |
Probing Simile Knowledge from Pre-trained Language Models (2022.acl-long)
Copied to clipboard
Weijie Chen, Yongzhu Chang, Rongsheng Zhang, Jiashu Pu, Guandan Chen, Le Zhang, Yadong Xi, Yijiang Chen, Chang Su
| Challenge: | Existing approaches to learn generic knowledge from a large corpus are time-consuming and labor-intensive. |
| Approach: | They propose a framework to probe simile knowledge from pre-trained language models to solve SI and SG tasks. |
| Outcome: | The proposed framework solves the SI and SG tasks in a simile triple completion task. |
Enhancing LLM Knowledge Learning through Generalization (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Continued pre-training on paraphrased data has shown empirical promise for enhancing knowledge acquisition, but this approach is costly and unreliable as it relies on external models or manual effort for rewriting. |
| Approach: | They propose formatting-based data augmentation which diversifies documents conveying the same knowledge by altering document formats rather than their content. |
| Outcome: | The proposed methods improve generalization to diverse paraphrased contexts and enhance pre-training and instruction tuning. |
Inter-Passage Verification for Multi-evidence Multi-answer QA (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing multi-answer question answering systems struggle to retrieve and synthesize a large number of evidence passages. |
| Approach: | They propose a multi-answer question answering framework that generates a large set of passages and then processes each passage individually to generate an initial high-recall but noisy answer set. |
| Outcome: | The proposed framework outperforms baselines on the QAMPARI and RoMQA datasets, achieving an average F1 score improvement of 11.17%. |
The Efficiency vs. Accuracy Trade-off: Optimizing RAG-Enhanced LLM Recommender Systems Using Multi-Head Early Exit (2025.acl-long)
Copied to clipboard
Huixue Zhou, Hengrui Gu, Zaifu Zhan, Xi Liu, Kaixiong Zhou, Yongkang Xiao, Mingfu Liang, Srinivas Prasad Govindan, Piyush Chawla, Jiyan Yang, Xiangfei Meng, Huayu Li, Buyun Zhang, Liang Luo, Wen-Yen Chen, Yiping Han, Bo Long, Rui Zhang, Tianlong Chen
| Challenge: | Existing frameworks for Large Language Models (LLMs) for Click-Through Rate prediction require a careful balance between computational efficiency and predictive accuracy. |
| Approach: | They propose a framework that integrates Retrieval-Augmented Generation with a novel multi-head early exit architecture to address both challenges. |
| Outcome: | The proposed framework reduces retrieval time while maintaining high model performance. |
Multiple Knowledge-Enhanced Interactive Graph Network for Multimodal Conversational Emotion Recognition (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Multimodal Emotion Recognition in Conversations models struggle due to lack of Common Sense Knowledge (CSK). |
| Approach: | They propose a multimodal approach to integrate multiple knowledge into the edge representations by integrating textual and visual CSK. |
| Outcome: | The proposed model outperforms state-of-the-art methods on two popular datasets. |
MuRAG: Multimodal Retrieval-Augmented Generator for Open Question Answering over Images and Text (2022.emnlp-main)
Copied to clipboard
| Challenge: | Pre-trained language models store a massive amount of world knowledge implicitly in their parameters, but large models often fail to encode information about rare entities and events. |
| Approach: | They propose a retrieval-augmented model which accesses an external non-parametric memory to augment language generation. |
| Outcome: | The proposed model outperforms existing models by 10-20% absolute on two datasets and under distractor and full-wiki settings. |
Progressive Re-ranking for Multimodal Retrieval-Augmented Generation via Curriculum Learning (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing approaches to improve retrieval performance of large language models are limited by static knowledge. |
| Approach: | They propose a multimodal re-ranking framework that combines curriculum learning with fine-grained reranking and multimodal section reassessment to improve CLIP-based visual coarse-grain retrieval. |
| Outcome: | The proposed framework achieves state-of-the-art answer accuracy and competitive retrieval performance on InfoSeek and Enc-VQA. |
Youling: an AI-assisted Lyrics Creation System (2020.emnlp-demos)
Copied to clipboard
Rongsheng Zhang, Xiaoxi Mao, Le Li, Lin Jiang, Lin Chen, Zhiwei Hu, Yadong Xi, Changjie Fan, Minlie Huang
| Challenge: | Recent studies have focused on a single pass of lyrics generation with little human intervention. |
| Approach: | They propose an AI-assisted lyrics creation system that supports one pass full-text generation and interactive generation modes. |
| Outcome: | The proposed system supports full-text generation and interactive generation modes . it also provides a revision module which enables users to revise undesired lyrics repeatedly. |
Improving Discriminative Capability of Reward Models in RLHF Using Contrastive Learning (2024.emnlp-main)
Copied to clipboard
Lu Chen, Rui Zheng, Binghai Wang, Senjie Jin, Caishuang Huang, Junjie Ye, Zhihao Zhang, Yuhao Zhou, Zhiheng Xi, Tao Gui, Qi Zhang, Xuanjing Huang
| Challenge: | Current methods rely on ranking losses to teach reward model to assess preferences, but they are susceptible to noise and ambiguous data, often failing to deeply understand human intentions. |
| Approach: | They propose a method that incorporates contrastive learning into the reward modeling process to enhance generalization and stabilize the reinforcement learning training process. |
| Outcome: | The proposed method enhances generalization of the reward model, stabilizes the reinforcement learning training process, and improves the final alignment with human preferences. |
GFT: From Imitation to Reward Fine-Tuning with Unbiased Group Advantages and Dynamic Coefficient Rectification (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing studies have demonstrated that supervised fine-tuning and reinforcement learning are effective in integrating knowledge injection with robust generalization. |
| Approach: | They propose a unified post-training framework that addresses intrinsic limitations of supervised fine-tuning and reinforcement learning. |
| Outcome: | The proposed framework surpasses SFT-based methods and yields policies that integrate more smoothly with subsequent RL training. |
Long-tail Relation Extraction via Knowledge Graph Embeddings and Graph Convolution Networks (N19-1)
Copied to clipboard
| Challenge: | Existing distance supervised relation extraction models for long-tail data are inadequate for many applications. |
| Approach: | They propose to leverage implicit relational knowledge among class labels and learn explicit relational knowing using graph convolution networks. |
| Outcome: | The proposed approach outperforms baselines for long-tail relations on a large-scale dataset. |
VRoPE: Rotary Position Embedding for Video Large Language Models (2025.emnlp-main)
Copied to clipboard
Zikang Liu, Longteng Guo, Yepeng Tang, Tongtian Yue, Junxian Cai, Kai Ma, Qingbin Liu, Xi Chen, Jing Liu
| Challenge: | Existing versions of Large Language Models (LLMs) lack a positional encoding strategy for video. |
| Approach: | They propose a new positional encoding method tailored for Video-LLMs that mitigates positional biases and ensures a more uniform distribution of spatial focus. |
| Outcome: | The proposed method outperforms existing versions of RoPE in video understanding and reasoning tasks. |
Analyzing Chain-of-thought Prompting in Black-Box Large Language Models via Estimated V-information (2024.lrec-main)
Copied to clipboard
| Challenge: | Chain-of-Thought (CoT) prompting and large language models (LLMs) have shown great potential in improving performance on challenging reasoning tasks. |
| Approach: | They propose a new metric which extends the concept of pointwise V-information to black-box models and quantifies label-relevant new information introduced by CoT prompting. |
| Outcome: | The proposed metric extends the concept of pointwise V-information to black-box models, quantifying label-relevant new information introduced by CoT prompting beyond pre-existing label information. |
TLSA: LLM-Guided Text-Label Space Alignment with Contrastive Learning for Generalized Category Discovery (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for generalized category discovery suffer from weak text–label alignment, inconsistent objectives across known and novel categories, and poor discrimination of semantically similar clusters. |
| Approach: | They propose a unified framework that enforces contrastive alignment between text and label representations within a shared semantic space. |
| Outcome: | The proposed framework outperforms state-of-the-art methods on four benchmark datasets. |
E-ConvRec: A Large-Scale Conversational Recommendation Dataset for E-Commerce Customer Service (2022.lrec-1)
Copied to clipboard
Meihuizi Jia, Ruixue Liu, Peiying Wang, Yang Song, Zexi Xi, Haobin Li, Xin Shen, Meng Chen, Jinhui Pang, Xiaodong He
| Challenge: | Recent research has focused on developing conversational recommendation system (CRS), which provides valuable recommendations to users through conversations. |
| Approach: | They construct an authentic Chinese dialogue dataset consisting of over 25k dialogues and 770k utterances, which contains user profile, product knowledge base, and multiple sequential real conversations between users and recommenders. |
| Outcome: | The proposed dataset contains user profile, product knowledge base, and multiple sequential real conversations between users and recommenders. |
Novel Relation Detection: Discovering Unknown Relation Types via Multi-Strategy Self-Supervised Learning (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing approaches to relation extraction can only recognize predefined relation types . new or out-of-scope relation types may continually emerge after the model is deployed . |
| Approach: | They propose a novel relation detection task that uses self-supervised learning to handle shallow semantic similarity problem. |
| Outcome: | The proposed method outperforms state-of-the-art methods on two datasets. |
SciAssess: Benchmarking LLM Proficiency in Scientific Literature Analysis (2025.findings-naacl)
Copied to clipboard
Hengxing Cai, Xiaochen Cai, Junhan Chang, Sihang Li, Lin Yao, Wang Changxin, Zhifeng Gao, Hongshuai Wang, Li Yongge, Mujie Lin, Shuwen Yang, Jiankun Wang, Mingjun Xu, Jin Huang, Xi Fang, Jiaxi Zhuang, Yuqi Yin, Yaqi Li, Changhong Chen, Zheng Cheng, Zifeng Zhao, Linfeng Zhang, Guolin Ke
| Challenge: | Existing benchmarks fail to adequately evaluate the proficiency of Large Language Models (LLMs) Existing standards do not cover the skills needed to evaluate LLMs in scientific literature analysis. |
| Approach: | They propose a benchmark to evaluate the proficiency of large language models in scientific literature analysis. |
| Outcome: | SciAssess evaluates 11 LLMs on multiple tasks across scientific fields. |
AgentV-RL: Scaling Reward Modeling with Agentic Verifier (2026.findings-acl)
Copied to clipboard
Jiazheng Zhang, Ziche Fu, Zhiheng Xi, Wenqing Jing, Mingxu Chai, Wei He, Guoqiang Zhang, Chenghao Fan, Chenxin An, Wenxiang Chen, Zhicheng Liu, Haojie Pan, Dingwei Zhu, Tao Gui, Qi Zhang, Xuanjing Huang
| Challenge: | Existing approaches to improve LLM reasoning are limited in complex domains and lack external grounding makes verifiers unreliable on computation-intensive tasks. |
| Approach: | They propose a framework that transforms reward modeling into a multi-turn, tool-augmented deliberative process. |
| Outcome: | The proposed framework surpasses state-of-the-art ORMs by 25.2% under parallel and sequential TTS. |
PFDial: A Structured Dialogue Instruction Fine-tuning Method Based on UML Flowcharts (2025.findings-acl)
Copied to clipboard
Ming Zhang, Yuhui Wang, Yujiong Shen, Tingyi Yang, Changhao Jiang, Yilong Wu, Shihan Dou, Qinhao Chen, Zhiheng Xi, Zhihao Zhang, Yi Dong, Zhen Wang, Zhihui Fei, Mingyang Wan, Tao Liang, Guojun Ma, Qi Zhang, Tao Gui, Xuanjing Huang
| Challenge: | Large Language Models (LLMs) have shown remarkable progress in dialogue and reasoning, but they struggle to solve strictly constrained dialogue tasks. |
| Approach: | They construct a dataset that contains 12,705 high-quality Chinese dialogue instructions from 440 flowcharts containing 5,055 process nodes. |
| Outcome: | The proposed model outperforms GPT-4o models on backward transitions and outperformed GPT-42 models on the same dataset. |
A Table-to-Text Framework with Heterogeneous Multidominance Attention and Self-Evaluated Multi-Pass Deliberation (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Table-to-text works have been widely applied in different domains, such as weather forecast and financial report generation. |
| Approach: | They propose a table-to-text approach on top of Self-evaluated multi-pass Generation and Heterogenous Multidominance Attention to explore the hierarchical structure. |
| Outcome: | The proposed method outperforms several SOTA methods quantitatively and qualitatively on three public datasets. |
TriSPrompt: A Hierarchical Soft Prompt Model for Multimodal Rumor Detection with Incomplete Modalities (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing multimodal rumor detection methods focus on learning joint modality representations from complete multimodal training data, rendering them ineffective in addressing the common occurrence of missing modalities in real-world scenarios. |
| Approach: | They propose a hierarchical soft prompt model TriSPrompt which integrates three types of prompts to effectively detect rumors in incomplete multimodal data. |
| Outcome: | The proposed model achieves an accuracy gain of over 13% compared to state-of-the-art models. |
What Makes an LLM a Good Optimizer? A Trajectory Analysis of LLM-Guided Evolutionary Search (2026.findings-acl)
Copied to clipboard
| Challenge: | Recent work has demonstrated the promise of orchestrating large language models (LLMs) within evolutionary and agentic optimization systems. |
| Approach: | They present a large-scale study of LLM-guided evolutionary search . they find strong LLMs behave as local refiners, producing frequent improvements . weaker LLM optimizers exhibit large semantic drift, they say . |
| Outcome: | The results highlight the importance of trajectory analysis for understanding and improving LLM-based optimization systems. |
AgentGym: Evaluating and Training Large Language Model-based Agents across Diverse Environments (2025.acl-long)
Copied to clipboard
Zhiheng Xi, Yiwen Ding, Wenxiang Chen, Boyang Hong, Honglin Guo, Junzhe Wang, Xin Guo, Dingwen Yang, Chenyang Liao, Wei He, Songyang Gao, Lu Chen, Rui Zheng, Yicheng Zou, Tao Gui, Qi Zhang, Xipeng Qiu, Xuanjing Huang, Zuxuan Wu, Yu-Gang Jiang
| Challenge: | Large language models (LLMs) are promising foundations to build generally-capable agents . however, the community lacks a unified interactive framework that covers diverse environments for comprehensive evaluation of agents. |
| Approach: | They propose a framework that features 7 real-world scenarios, 14 environments, and 89 tasks for unified, real-time, and concurrent agent interaction. |
| Outcome: | The proposed framework features 7 real-world scenarios, 14 environments, and 89 tasks for unified, real-time, and concurrent agent interaction. |
An Industry Evaluation of Embedding-based Entity Alignment (2020.coling-industry)
Copied to clipboard
| Challenge: | Knowledge graphs (KGs) are increasingly important in various applications such as question answering and search engines. |
| Approach: | They propose to use a supervised learning environment with unbiased seed mappings for training and validation to evaluate alignment methods in an industrial context. |
| Outcome: | The proposed methods are evaluated in an industrial context and are compared with DBpedia and Wikidata benchmarks. |
OntoEA: Ontology-guided Entity Alignment via Joint Knowledge Graph Embedding (2021.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for aligning knowledge graph entities ignore the ontology which contains critical meta information such as classes and membership relationships with entities. |
| Approach: | They propose an ontology-guided method where KGs and ontologies are jointly embedded. |
| Outcome: | Extensive experiments on seven public and industrial benchmarks show the ontology-guided method performs well and is cost-effective. |
Contextualized End-to-End Neural Entity Linking (2020.aacl-main)
Copied to clipboard
| Challenge: | a proposed entity linking model that disjointly applies MD and ED from the same contextualized BERT embeddings is able to generalize better. |
| Approach: | They propose an entity linking (EL) model that jointly learns mention detection (MD) and entity disambiguation (ED) they propose to use task-specific heads on top of shared BERT contextualized embeddings to learn MD and ED. |
| Outcome: | The proposed model achieves state-of-the-art results across a standard EL dataset and under a setting where hand-crafted candidate sets are not available. |
PRGC: Potential Relation and Global Correspondence Based Joint Relational Triple Extraction (2021.acl-long)
Copied to clipboard
Hengyi Zheng, Rui Wen, Xi Chen, Yifan Yang, Yunyan Zhang, Ziheng Zhang, Ningyu Zhang, Bin Qin, Xu Ming, Yefeng Zheng
| Challenge: | Recent methods for extracting entities and relations from unstructured texts suffer from limitations, such as redundancy of relation prediction and inefficiency. |
| Approach: | They propose a joint relational triple extraction framework based on Potential Relation and Global Correspondence (PRGC) they propose overlapping triples for relation prediction and relation-relational alignment . |
| Outcome: | The proposed framework achieves state-of-the-art performance on public benchmarks with higher efficiency and consistent performance gain on complex scenarios of overlapping triples. |
Editing the Moving World: Model Editing for Video LLMs (2026.acl-long)
Copied to clipboard
Qian Zhang, Xinye Li, Xiaokai Wu, Junhao Xu, Zhanyue Qin, Qingbin Liu, Junxian Cai, Xi Chen, Bolin Zhang, Zhiying Tu, Dianhui Chu, Xiaoyan Yu, Dianbo Sui
| Challenge: | Existing models for knowledge editing focus on knowledge-level or static visual domains, overlooking dynamic semantics. |
| Approach: | They propose a benchmark for modeling large language models using six representative models . they analyze the strengths and limitations of existing models and identify new directions . |
| Outcome: | The proposed benchmark extends existing models from static modalities to dynamic video scenarios. |
Imperfect also Deserves Reward: Multi-Level and Sequential Reward Modeling for Better Dialog Management (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing research on taskoriented dialog systems mainly includes pipeline and end-to-end methods due to its non-differentiable nature. |
| Approach: | They propose a multi-level reward modeling approach that factorizes a reward into a three-level hierarchy: domain, act, and slot. |
| Outcome: | The proposed approach significantly improves performance and speed of training in a wide range of dialog systems. |
Detoxifying Large Language Models via Knowledge Editing (2024.acl-long)
Copied to clipboard
Mengru Wang, Ningyu Zhang, Ziwen Xu, Zekun Xi, Shumin Deng, Yunzhi Yao, Qishen Zhang, Linyi Yang, Jindong Wang, Huajun Chen
| Challenge: | Existing methods to detoxify Large Language Models (LLMs) are limiting, but knowledge editing can be effective. |
| Approach: | They propose a baseline method to detoxify Large Language Models (LLMs) they propose supervised fine-tuning and reinforcement learning from human feedback (RLHF) |
| Outcome: | The proposed method reduces toxicity of large language models with one instance of tuning . it reduces the toxicity, while minimizing the toxins, the authors show . |
All You May Need for VQA are Image Captions (2022.naacl-main)
Copied to clipboard
| Challenge: | Visual Question Answering (VQA) has benefited from increasingly sophisticated models, but has not enjoyed the same level of engagement in terms of data creation. |
| Approach: | They propose a method that automatically derives VQA examples at volume by leveraging existing image-caption annotations combined with neural models for textual question generation. |
| Outcome: | The proposed method improves state-of-the-art zero-shot accuracy by double digits and achieves robustness that lacks in the same model trained on human-annotated VQA data. |
Task-wrapped Continual Learning in Task-Oriented Dialogue Systems (2025.findings-naacl)
Copied to clipboard
| Challenge: | Continual learning is vital for task-oriented dialogue systems (ToDs), but its performance is limited by training separate adapters for each task, preventing global knowledge sharing. |
| Approach: | They propose a framework that employs task-wrapped Adapters to learn global and task-specific information through parameter sharing. |
| Outcome: | The proposed framework outperforms AdapterCL in 37 domains while using only 46% of the parameters. |
OmniThink: Expanding Knowledge Boundaries in Machine Writing through Thinking (2025.emnlp-main)
Copied to clipboard
Zekun Xi, Wenbiao Yin, Jizhan Fang, Jialong Wu, Runnan Fang, Yong Jiang, Pengjun Xie, Fei Huang, Huajun Chen, Ningyu Zhang
| Challenge: | Recent advances in Large Language Models (LLMs) have demonstrated remarkable progress in machine writing such as open domain long-form generation. |
| Approach: | They propose a slow-thinking machine writing framework that emulates the human-like process of iterative expansion and reflection to improve the knowledge density of generated articles. |
| Outcome: | The proposed framework improves the knowledge density of generated articles without compromising metrics such as coherence and depth. |
STELLA: A Multimodal LLM for Protein Functional Annotation via Unified Sequence-Structure Encoding (2026.findings-acl)
Copied to clipboard
Hongwang Xiao, Wenjun Lin, Xi Chen, Hui Wang, Kai Chen, Jiashan Li, Yuancheng Sun, Sicheng Dai, Boya Wu, Qiwei Ye
| Challenge: | a multimodal protein language model (LLM) integrates sequence, structure, and function into functional annotation. |
| Approach: | They propose a multimodal protein language model that synergistically aligns bimodal representations with the textual modality to advance protein functional annotation. |
| Outcome: | The proposed model synergizes bimodal representations with the textual modality to advance protein functional annotation. |
Sparse Adapter Fusion for Continual Learning in NLP (2026.eacl-long)
Copied to clipboard
| Challenge: | Existing methods for learning continuously face challenges such as inefficient parameter reuse across tasks and catastrophic forgetting when tasks are dissimilar. |
| Approach: | They propose a Sparse Adapter Fusion Method which dynamically fuses old and new adapters to address these challenges. |
| Outcome: | The proposed method outperforms state-of-the-art methods while utilizing less than 60% of the parameters. |
LLaST: Improved End-to-end Speech Translation System Leveraged by Large Language Models (2024.findings-acl)
Copied to clipboard
| Challenge: | ***LLaST*** is a framework for building high-performance Large Language model based Speech-to-text Translation systems. |
| Approach: | They propose a framework for building high-performance Large Language model based Speech-to-text Translation systems. |
| Outcome: | The proposed model outperforms the CoVoST-2 benchmark and showcases exceptional scaling capabilities powered by LLMs. |
Prototype Conditioned Generative Replay for Continual Learning in NLP (2025.naacl-long)
Copied to clipboard
| Challenge: | Generative replay methods that rely on a single task-specific token or prompt often fail to generate pseudo-samples that accurately reflect the true data distribution. |
| Approach: | They propose a Prototype Conditioned Generative Replay method which incorporates task-level statistics into a prototyping process. |
| Outcome: | The proposed method outperforms state-of-the-art (SOTA) methods on two different scenarios. |
Efficient Self-Evaluation for Diffusion Language Models via Sequence Regeneration (2026.acl-long)
Copied to clipboard
| Challenge: | Non-sequential and bidirectional nature of diffusion large language models makes direct likelihood-based self-evaluation challenging. |
| Approach: | They propose a self-evaluation confidence quantification method for diffusion large language models that quantifies confidence by computing the probability of regenerating tokens in the entire generated sequence, given the full context. |
| Outcome: | The proposed method is correlated with semantic coherence and answer accuracy. |
Summarizing Chinese Medical Answer with Graph Convolution Networks and Question-focused Dual Attention (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Existing approaches to generate answer summarization for medical questions are not straightforward to apply to the medical domain. |
| Approach: | They propose an approach that utilizes graph convolution networks and question-focused dual attention for Chinese medical answer summarization. |
| Outcome: | The proposed model generates more coherent and informative summaries compared with baseline models. |
Can We Edit Multimodal Large Language Models? (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to edit multimodal models have been used to incrementally infuse a language model with a new set of facts. |
| Approach: | They construct a benchmark for editing multimodal Large Language Models and establish metrics for evaluation. |
| Outcome: | The proposed benchmarks show that editing multimodal models is not as difficult as editing single-modal models. |
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. |
BOLT: Benchmarking Open-World Learning for Text Classification (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing benchmarks focus on out-of-distribution (OOD) detection while overlooking broader challenges such as the discovery of novel categories. |
| Approach: | They propose a unified Benchmark and evaluation toolkit supporting Open-world learning for text classification. |
| Outcome: | The proposed methods overfit training distributions and struggle to generalize to unseen classes. |
Query-Focused Retrieval Heads Improve Long-Context Reasoning and Re-ranking (2025.emnlp-main)
Copied to clipboard
| Challenge: | Recent work has identified retrieval heads as a subset of attention heads responsible for retrieving salient information in long-context language models. |
| Approach: | They introduce a retrieval head that uses attention scores to enhance retrieval from long context . they use QRRetriever to select the most relevant parts with the highest retrieval scores . |
| Outcome: | The proposed retrieval heads outperform other retrieval-based retrieval retrievers on BEIR benchmarks. |
Explanation in the Era of Large Language Models (2024.naacl-tutorials)
Copied to clipboard
| Challenge: | Explanation has long been a part of communication, where humans use language to elucidate each other and transmit information about mechanisms of events. |
| Approach: | They review the opportunities and challenges of explanations in the era of large language models and examine how they can be used to generate explanations. |
| Outcome: | The proposed methods are based on the models of large language models (LLMs) and their opaque nature. |
How to Determine the Most Powerful Pre-trained Language Model without Brute Force Fine-tuning? An Empirical Survey (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Transferability estimation has been a topic of great interest in computer vision fields . a lack of a comprehensive comparison between these estimation methods is a problem . |
| Approach: | They conduct a thorough survey of existing methods to find the most suitable model . they also outline difficulties of consideration of training details and applicability to text generation . |
| Outcome: | The proposed methods perform well with superiorities in effectiveness and efficiency. |
Dream to Chat: Model-based Reinforcement Learning on Dialogues with User Belief Modeling (2025.findings-emnlp)
Copied to clipboard
Yue Zhao, Xiaoyu Wang, Dan Wang, Zhonglin Jiang, Qingqing Gu, Teng Chen, Ningyuan Xi, Jinxian Qu, Yong Chen, Luo Ji
| Challenge: | a framework for constructing dialogue world models for natural language tasks is currently lacking. |
| Approach: | They propose a framework that can be used to train a dialogue world model. |
| Outcome: | The proposed framework can predict future utterances and user beliefs . it can achieve state-of-the-art performance on emotion classification and sentiment identification . |
Towards Realistic Low-resource Relation Extraction: A Benchmark with Empirical Baseline Study (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing approaches to extract relational facts from text are limited in their ability to learn from limited labeled data. |
| Approach: | They propose to use prompt-based methods with few-shot labeled data to evaluate performance . data augmentation technologies and self-training are also proposed to generate more labeles in-domain data. |
| Outcome: | The proposed methods perform well in low-resource settings with 8 relation extraction datasets. |
Finding Influential Instances for Distantly Supervised Relation Extraction (2022.coling-1)
Copied to clipboard
| Challenge: | Distant supervision models suffer from high label noise and are not reliable for DS. |
| Approach: | They propose a model-agnostic instance sampling method for relation extraction (RE) by influence function, namely REIF. |
| Outcome: | The proposed method reduces the computational complexity from O(mn) to O(1), with analyzing its robustness on the selected sampling function. |
All That Glisters Is Not Gold: A Benchmark for Reference-Free Counterfactual Financial Misinformation Detection (2026.acl-long)
Copied to clipboard
Yuechen Jiang, Zhiwei Liu, Yupeng Cao, Yueru He, Ziyang Xu, Chen Xu, Zhiyang Deng, Prayag Tiwari, Xi Chen, Alejandro Lopez-Lira, Jimin Huang, Junichi Tsujii, Sophia Ananiadou
| Challenge: | RFC-Bench evaluates large language models on financial misinformation under realistic news . current models struggle to maintain coherent belief states without external grounding, study finds . |
| Approach: | They propose a benchmark for evaluating large language models on financial misinformation under realistic news. |
| Outcome: | The proposed model performs better when context is available, while reference-free settings expose significant weaknesses. |
Reward Modeling Requires Automatic Adjustment Based on Data Quality (2024.findings-emnlp)
Copied to clipboard
Binghai Wang, Rui Zheng, Lu Chen, Zhiheng Xi, Wei Shen, Yuhao Zhou, Dong Yan, Tao Gui, Qi Zhang, Xuanjing Huang
| Challenge: | Reinforcement Learning from Human Feedback (RLHF) is a method for aligning language models with human values. |
| Approach: | They propose a method that automatically adjusts reward modeling based on data quality . they use preference data to train a reward model that is more aligned with human values . |
| Outcome: | The proposed method stabilizes reward model training and significantly improves alignment performance on human preference datasets. |
ScEdit: Script-based Assessment of Knowledge Editing (2025.findings-acl)
Copied to clipboard
Xinye Li, Zunwen Zheng, Qian Zhang, Dekai Zhuang, Jiabao Kang, Liyan Xu, Qingbin Liu, Xi Chen, Zhiying Tu, Dianhui Chu, Dianbo Sui
| Challenge: | Knowledge Editing (KE) has gained increasing attention, yet current evaluation frameworks do not integrate KE into real-world application scenarios. |
| Approach: | They propose a script-based benchmark which encompasses both counterfactual and temporal edits and integrates token-level and text-level evaluation methods. |
| Outcome: | The proposed method combines token-level and text-level evaluation methods with a new fact-based evaluation framework. |
AutoSchemaKG: Autonomous Knowledge Graph Construction through Dynamic Schema Induction from Web-Scale Corpora (2026.acl-long)
Copied to clipboard
Jiaxin Bai, Wei Fan, Qi Hu, Qing Zong, Chunyang Li, Hong Ting Tsang, Hongyu Luo, Yauwai Yim, Haoyu Huang, Xiao Zhou, Feng Qin, Tianshi Zheng, Xi Peng, Xin Yao, Huiwen Yang, Leijie Wu, JI Yi, Gong Zhang, Renhai Chen, Yangqiu Song
| Challenge: | Existing knowledge graph construction frameworks require predefined schemas, limiting their scalability and domain coverage. |
| Approach: | They propose a framework for fully autonomous knowledge graph construction that eliminates the need for predefined schemas. |
| Outcome: | The proposed framework outperforms state-of-the-art models on multi-hop QA tasks and enhances LLM factuality. |
Not All Directions Matter: Towards Structured and Task-Aware Low-Rank Model Adaptation (2026.acl-long)
Copied to clipboard
Xi Xiao, Chenrui Ma, Yunbei Zhang, Chen Liu, Zhuxuanzi Wang, Yanshu Li, Lin Zhao, Guosheng Hu, Tianyang Wang, Hao Xu
| Challenge: | Low-Rank Adaptation (LoRA) is a key parameter-efficient fine-tuning method . however, its effectiveness is hampered by semantic drift and structural incoherence . |
| Approach: | They propose a low-rank Adaptation framework that tackles semantic drift and structural incoherence by pruning task-irrelevant directions. |
| Outcome: | Experiments on large language models, vision models, and vision models show that the proposed framework outperforms LoRA and advanced dynamic rank allocation and sparsity-based methods. |
Optimal Neural Program Synthesis from Multimodal Specifications (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for multimodal program synthesis combine noisy signals from the user with hard constraints on the program’s behavior. |
| Approach: | They propose an optimal neural synthesis approach where the goal is to find a program that satisfies user-provided constraints while also maximizing the program’s score with respect to a neural model. |
| Outcome: | The proposed approach outperforms prior state-of-the-art methods in terms of accuracy and efficiency and finds model-optimal programs more frequently. |
SCCD: A Session-based Dataset for Chinese Cyberbullying Detection (2025.coling-main)
Copied to clipboard
| Challenge: | Existing work on cyberbullying detection in Chinese is underdeveloped due to the lack of comprehensive and reliable datasets. |
| Approach: | They propose to use Chinese social media sessions to analyze Chinese cyberbullying content to improve the quality of annotations. |
| Outcome: | The proposed dataset shows that it performs better than existing methods on Weibo and a major social media platform. |
SMASH: Improving SMAll Language Models’ Few-SHot Ability with Prompt-Based Distillation (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Large-scale language models with prompts have shown remarkable performance on few-shot learning. |
| Approach: | They propose an approach to improve SMAll language models’ few-SHot ability by training on intermediate tasks before prompt-based fine-tuning on downstream tasks. |
| Outcome: | The proposed model improves on sentence-pair and sentiment classification tasks by training on intermediate tasks before fine-tuning on downstream tasks. |
PromptFix: Few-shot Backdoor Removal via Adversarial Prompt Tuning (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing studies have shown that pre-trained language models can be backdoored such that model behavior is manipulated when trigger tokens are presented. |
| Approach: | They propose a backdoor mitigation strategy for NLP models via adversarial prompt-tuning in few-shot settings that uses two extra sets of soft tokens which approximate the trigger and counteract it respectively. |
| Outcome: | The proposed method keeps model parameters intact and only utilizes two extra sets of soft tokens which approximate the trigger and counteract it respectively. |
Label-Free Distant Supervision for Relation Extraction via Knowledge Graph Embedding (D18-1)
Copied to clipboard
| Challenge: | Existing methods to generate large scale labeled data for relation extraction produce noisy relation labels when there are multiple relationships between entities. |
| Approach: | They propose a method which assumes that a pair of entities appears in a Knowledge Graph and trains a relation classifier. |
| Outcome: | The proposed method performs well in the current distant supervision dataset. |
MultiFinBen: Benchmarking Large Language Models for Multilingual and Multimodal Financial Application (2026.acl-long)
Copied to clipboard
Xueqing Peng, Lingfei Qian, Yan Wang, Ruoyu Xiang, Yueru He, Yang Ren, Mingyang Jiang, Vincent Jim Zhang, Yuqing Guo, Jeff Zhao, Huan He, Yi Han, Yun Feng, Yuechen Jiang, Yupeng Cao, Haohang Li, Yangyang Yu, Xiaoyu Wang, Penglei Gao, Shengyuan Lin, Keyi Wang, Shanshan Yang, Yilun Zhao, Zhiwei Liu, Peng Lu, Jerry Huang, Suyuchen Wang, Triantafillos Papadopoulos, Polydoros Giannouris, Efstathia Soufleri, Nuo Chen, Zhiyang Deng, Heming Fu, Yijia Zhao, Mingquan Lin, Meikang Qiu, Kaleb E Smith, Arman Cohan, Xiao-Yang Liu, Jimin Huang, Guojun Xiong, Alejandro Lopez-Lira, Xi Chen, Junichi Tsujii, Jian-Yun Nie, Sophia Ananiadou, Qianqian Xie
| Challenge: | Existing evaluations of LLMs in finance are text-only, monolingual, and largely saturated by current models. |
| Approach: | They propose a multilingual and multimodal benchmark for evaluating LLMs in real financial contexts. |
| Outcome: | The first expert-annotated multilingual and multimodal benchmark is released . it evaluates 21 leading LLMs and shows they perform better in multilingual settings . |
GenDis: Generative-Discriminative Dual-View Co-Training for Generalized Category Discovery (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods rely on one-hot discriminative supervision, leading to overfitting on seen classes and poor generalization to unseen ones. |
| Approach: | They propose a Generative–Discriminative Dual-View Co-Training framework that unifies discriminative classification and semantic label generation within an LLM. |
| Outcome: | The proposed framework outperforms existing methods on five benchmarks on the generalized category discovery (GCD) task. |
Adversarial Preference Learning for Robust LLM Alignment (2025.findings-acl)
Copied to clipboard
Yuanfu Wang, Pengyu Wang, Chenyang Xi, Bo Tang, Junyi Zhu, Wenqiang Wei, Chen Chen, Chao Yang, Jingfeng Zhang, Chaochao Lu, Yijun Niu, Keming Mao, Zhiyu Li, Feiyu Xiong, Jie Hu, Mingchuan Yang
| Challenge: | Modern language models rely on Reinforcement Learning from Human Feedback (RLHF) to encourage safe behaviors, but they remain vulnerable to adversarial attacks due to three key limitations: (1) the inefficiency and high cost of human annotation; (2) the vast diversity of potential adversarials; and (3) the risk of feedback bias and reward hacking. |
| Approach: | They propose an iterative adversarial training method that incorporates three key innovations to address these challenges. |
| Outcome: | Experiments on Mistral-7B-Instruct-v0.3 show that the proposed method significantly enhances robustness and reduces harmful outputs from 5.88% to 0.43%. |
ORTicket: Let One Robust BERT Ticket Transfer across Different Tasks (2024.lrec-main)
Copied to clipboard
| Challenge: | Pretrained language models are susceptible to subtle perturbations and require multiple adversarial training during fine-tuning to improve their robustness. |
| Approach: | They propose a novel adversarial defense method ORTicket that fine-tunes a model for downstream tasks. |
| Outcome: | The proposed method achieves comparable robustness to other defense methods while maintaining the efficiency of fine-tuning. |
AJ-Bench: Benchmarking Agent-as-a-Judge for Environment-Aware Evaluation (2026.findings-acl)
Copied to clipboard
Wentao Shi, Yu Wang, Yuyang Zhao, Yuxin Chen, Fuli Feng, Xueyuan Hao, Xi Su, Qi GU, Hui Su, Xunliang Cai, Xiangnan He
| Challenge: | Existing approaches to verify agent behaviors in complex environments rely on rule-based verifiers or LLM-as-a-Judge models. |
| Approach: | They propose a benchmark to evaluate Agent-as-a-Judge across three domains . the benchmark covers search, data systems, and graphical user interfaces - with 155 tasks and 516 trajectories . |
| Outcome: | The proposed benchmark outperforms existing benchmarks in search, data systems, and GUI domains while revealing open challenges in agent-based verification. |
Class Lifelong Learning for Intent Detection via Structure Consolidation Networks (2023.findings-acl)
Copied to clipboard
Qingbin Liu, Yanchao Hao, Xiaolong Liu, Bo Li, Dianbo Sui, Shizhu He, Kang Liu, Jun Zhao, Xi Chen, Ningyu Zhang, Jiaoyan Chen
| Challenge: | Existing intent detection models can only handle predefined intent classes in the offline environment. |
| Approach: | They propose a method that continually learns new intent classes from new data . structure-based retrospection and contrastive knowledge distillation are used to solve these problems . |
| Outcome: | The proposed method outperforms existing models on three benchmarks. |
Exploring Deductive and Inductive Reasoning Capabilities of Large Language Models in Procedural Planning (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Deductive and inductive reasoning are fundamental components of human cognition . authors present a benchmark to assess their performance in procedural planning . |
| Approach: | They propose a benchmark to assess the deductive and inductive reasoning abilities of LLMs . they propose IMSE to enable LLM to generate multiple similar procedural plans . |
| Outcome: | The proposed method improves inductive reasoning abilities of LLMs, the authors show . they show that LLM models show excellent deductive reasoning capabilities but suboptimal inductive performance. |
ActorMind: Emulating Human Actor Reasoning for Speech Role-Playing (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing work on role-playing focuses on textual modalities, neglecting speech . et al., 2025) show that speech role-players can generate spontaneous responses with personalized traits based on the context. |
| Approach: | They propose a framework that allows models to deliver spontaneous responses with personalized verbal traits based on their role, scene, and spoken dialogue. |
| Outcome: | The proposed framework enhances speech role-playing by generating spontaneous responses with personalized traits based on their role, scene, and spoken dialogue. |
EasyEdit: An Easy-to-use Knowledge Editing Framework for Large Language Models (2024.acl-demos)
Copied to clipboard
Peng Wang, Ningyu Zhang, Bozhong Tian, Zekun Xi, Yunzhi Yao, Ziwen Xu, Mengru Wang, Shengyu Mao, Xiaohan Wang, Siyuan Cheng, Kangwei Liu, Yuansheng Ni, Guozhou Zheng, Huajun Chen
| Challenge: | Large Language Models (LLMs) suffer from knowledge cutoff or fallacy issues, which means they are unaware of unseen events or generate text with incorrect facts owing to outdated/noisy data. |
| Approach: | They propose an easy-to-use knowledge editing framework for Large Language Models that allows users to easily edit updated knowledge and adjust undesired behavior while minimizing the impact on unrelated inputs. |
| Outcome: | The proposed framework surpasses traditional fine-tuning in terms of reliability and generalization. |
Conan-Embedding-v2: Training an LLM from Scratch for Text Embeddings (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies use LoRA to fine-tune existing LLMs, but this is limited by the data and training gap between them and embedding models. |
| Approach: | They propose a new 1.4B-parameter LLM trained from scratch and fine-tuned as a text embedder that integrates embeddings across different languages. |
| Outcome: | The proposed model improves performance on the Massive Text Embedding Benchmark (MTEB) and Chinese MTEB (May 19, 2025). |
Benchmarking Multimodal Regex Synthesis with Complex Structures (2020.acl-main)
Copied to clipboard
| Challenge: | Existing datasets for regex generation from natural language are limited in complexity . Existing regex synthesis datasets are simple and the language used to describe them is not diverse . |
| Approach: | They propose a dataset for regex generation from natural language that generates regexes using a probabilistic grammar and pre-defined macros. |
| Outcome: | The proposed dataset is compared to existing datasets for regex generation from natural language . it generates the regexes using a probabilistic grammar with pre-defined macros observed from real-world StackOverflow posts. |
Debate-Feedback: A Multi-Agent Framework for Efficient Legal Judgment Prediction (2025.naacl-short)
Copied to clipboard
| Challenge: | Comparative experiments show that our model outperforms several general-purpose and domain-specific legal models. |
| Approach: | They propose a legal judgment prediction model that integrates LLMs with argumentative reasoning techniques to simulate the debate phase of real courtroom trials. |
| Outcome: | The proposed model outperforms several general-purpose and domain-specific legal models and offers a dynamic reasoning process. |
BFS-Prover: Scalable Best-First Tree Search for LLM-based Automatic Theorem Proving (2025.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to theorem proving in large language models rely on value functions and/or Monte Carlo Tree Search (MCTS), but the potential of simpler methods like Best-First Tree Search remains underexplored. |
| Approach: | They propose a scalable expert iteration framework that implements strategic data filtering at each expert iteration round, excluding problems solvable via beam search node expansion to focus on harder cases. |
| Outcome: | The proposed framework achieves a state-of-the-art score of 72.95 on the MiniF2F test set and challenges the perceived necessity of complex tree search methods. |
UER: An Open-Source Toolkit for Pre-training Models (D19-3)
Copied to clipboard
Zhe Zhao, Hui Chen, Jinbin Zhang, Xin Zhao, Tao Liu, Wei Lu, Xi Chen, Haotang Deng, Qi Ju, Xiaoyong Du
| Challenge: | Existing work on pre-training models have shown that it is important to use a framework to deploy various pre- training models efficiently. |
| Approach: | They propose an assemble-on-demand pre-training toolkit that assembles pre-trained models on demand and encapsulates them with rich modules. |
| Outcome: | The proposed framework can reproduce state-of-the-art models or develop models that remain unexplored. |
Enhancing Efficiency and Exploration in Reinforcement Learning for LLMs (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches allocate an equal number of rollouts to all questions during the RL process, which is inefficient. |
| Approach: | They propose a mechanism for dynamically allocating rollout budgets based on the difficulty of the problems, enabling more efficient RL training. |
| Outcome: | The proposed model improves response precision while preserving exploratory ability to uncover potential correct pathways. |
ReasonRec: A Reasoning-Augmented Multimodal Agent for Unified Recommendation (2026.findings-acl)
Copied to clipboard
Yihua Zhang, Mingfu Liang, Jiyan Yang, Rong Jin, Wen-Yen Chen, Yiping Han, Huayu Li, Buyun Zhang, Liang Luo, Luke Simon, Sijia Liu, Tianlong Chen, Xi Liu
| Challenge: | Recent advances in multimodal recommenders lack explicit reasoning and self-awareness of uncertainty. |
| Approach: | They propose a reasoning-augmented multimodal agent structured around a three-stage explicit reasoning pipeline. |
| Outcome: | The proposed agent improves ranking metrics and performance on four standard recommendation tasks across five real-world datasets. |
Discourse Structure-Aware Prefix for Generation-Based End-to-End Argumentation Mining (2024.findings-acl)
Copied to clipboard
| Challenge: | Recent advances in AM models overlook the integration of supplementary discourse structure information, resulting in suboptimal outcomes. |
| Approach: | They propose a framework which generates discourse structure-aware prefixes for each layer of the generation model. |
| Outcome: | The proposed framework achieves state-of-the-art performance on two AM benchmarks. |
Pragmatic Inference Chain (PIC) Improving LLMs’ Reasoning of Authentic Implicit Toxic Language (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have shown that LLMs can detect toxicity by using a variety of inference-intensive tasks, such as understanding humour and metaphors. |
| Approach: | They propose a new method to prompt LLMs to identify toxic language using a set of online data that are verified by human annotators. |
| Outcome: | The proposed method significantly improves the success rate of GPT-4o, Llama-3.1-70B-Instruct, DeepSeek-v2.5, and DeepSeq-v3 in identifying implicit toxic language compared to five baseline prompts, such as CoT and rule-based baselines. |
Incorporating Review-missing Interactions for Generative Explainable Recommendation (2025.coling-main)
Copied to clipboard
| Challenge: | Existing models of explainable recommendation use user reviews as ground truths, but in practice, a large amount of users may not leave reviews after purchasing items. |
| Approach: | They propose to incorporate user preferences into explainable recommender models by leveraging generative models to predict the missing reviews and then training the model based on all the predicted and original reviews. |
| Outcome: | The proposed model improves the explanation quality on three publicly available datasets. |
Schema-adaptable Knowledge Graph Construction (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing Knowledge Graph Construction (KGC) tasks rely on static information extraction with a closed set of pre-defined schemas. |
| Approach: | They propose a static knowledge Graph Construction task that extracts entity, relation, and event based on dynamically changing schema graph without retraining. |
| Outcome: | The proposed system outperforms existing methods but still has room for improvement . it can extract entity, relation, and event based on dynamically changing schema graph without re-training . |
GBV-SQL: Guided Generation and SQL2Text Back-Translation Validation for Multi-Agent Text2SQL (2026.acl-long)
Copied to clipboard
| Challenge: | Existing frameworks for Text2SQL generation still have a critical semantic gap . a dedicated validator translates generated SQL back into natural language and checks whether its logic is aligned with the original question. |
| Approach: | They propose a framework that introduces Guided Generation with SQL2Text Back-translation Validation . dedicated validator translates generated SQL back into natural language and checks whether logic is aligned with original question . |
| Outcome: | The proposed framework achieves 63.23% execution accuracy on the BIRD benchmark and 90.42% on repaired BIDR dev. |
Parrot: A Training Pipeline Enhances Both Program CoT and Natural Language CoT for Reasoning (2025.emnlp-main)
Copied to clipboard
Senjie Jin, Lu Chen, Zhiheng Xi, Yuhui Wang, Sirui Song, Yuhao Zhou, Xinbo Zhang, Peng Sun, Hong Lu, Tao Gui, Qi Zhang, Xuanjing Huang
| Challenge: | Existing work focuses on enabling models to generate natural language chain-of-thought rationales or leverage executable and verifiable code, such as Python. |
| Approach: | They propose a novel training pipeline that integrates sequential P-CoT and N-Co T generation and a subtask hybrid training strategy to facilitate natural language transferability. |
| Outcome: | The proposed training pipeline improves both N-CoT and P-Co T performance over the RL baseline. |
Beyond Read-Only: Crafting a Comprehensive Chinese Text-to-SQL Dataset for Database Manipulation and Query (2024.findings-naacl)
Copied to clipboard
| Challenge: | Current research focuses mainly on read operations and ignores other aspects of database operations such as create, update, and delete operations. |
| Approach: | They propose a large-scale cross-domain single-table CRUD operations Chinese Text-to-SQL dataset . the dataset contains 10,000 question/SQl pairs involving 625 tables from different domains . |
| Outcome: | The proposed method achieves 67.08% and 83.8% exact set matching accuracy under read and delete operations, but only 49.6% and 61.8% under create and update operations. |
Crossmodal-3600: A Massively Multilingual Multimodal Evaluation Dataset (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies on multilingual image captioning have been hampered by a lack of high-quality evaluation datasets. |
| Approach: | They present a dataset of 3600 images annotated with human-generated captions in 36 languages. |
| Outcome: | The proposed dataset shows that it is feasible to build multilingual image captioning models trained on machine-translated data. |
Ambiguity Awareness Optimization: Towards Semantic Disambiguation for Direct Preference Optimization (2025.emnlp-main)
Copied to clipboard
| Challenge: | Direct Preference Optimization (DPO) is a widely used reinforcement learning from human feedback (RLHF) method across various domains. |
| Approach: | They propose an approach that automatically re-weights ambiguous content to reduce ambiguities by calculating semantic similarity from preference pairs. |
| Outcome: | The proposed approach outperforms state-of-the-art approaches in performance across multiple model scales and widely adopted benchmark datasets. |
BioGraphia: A LLM-Assisted Biological Pathway Graph Annotation Platform (2025.emnlp-demos)
Copied to clipboard
| Challenge: | Existing methods for obtaining pathway information from biomedical literature rely on simplifying assumptions that limit their ability to capture true complexity of biological reactions. |
| Approach: | They propose a web-based platform to facilitate collaborative pathway graph annotation. |
| Outcome: | The platform supports multi-user collaboration with real-time monitoring, curation, and interactive pathway graph visualization. |
ComRAG: Retrieval-Augmented Generation with Dynamic Vector Stores for Real-time Community Question Answering in Industry (2025.acl-industry)
Copied to clipboard
| Challenge: | Existing methods for Community Question Answering (CQA) focus on static knowledge, limiting their applicability to real-world scenarios. |
| Approach: | They propose a retrieval-augmented generation framework for real-time industrial CQA that integrates static knowledge with dynamic historical QA pairs via a centroid-based memory mechanism. |
| Outcome: | The proposed framework outperforms baselines on three industrial CQA datasets and achieves 25.9% improvement in vector similarity, reducing latency by 8.7%–23.3%, and lowering chunk growth from 20.23% to 2.06% over iterations. |
Decoding Scientific Experimental Images: The SPUR Benchmark for Perception, Understanding, and Reasoning (2026.acl-long)
Copied to clipboard
Junpeng Ding, Zichen Tang, Haihong E, Mengyuan Ji, Yang Liu, Haolin Tian, Haiyang Sun, Pengqi Sun, Yang Xu, Yichen Liu, Haocheng Gao, Zijie Xi, Ruomeng Jiang, Peizhi Zhao, Rongjin Li, Yuanze Li, Jiacheng Liu, Zhongjun Yang, Jintong Chen, Siying Lin
| Challenge: | Xu and Peng, 2025) . . SPUR is a comprehensive benchmark for scientific experimental image perception, understanding, and reasoning, comprising 4,264 question-answering (QA) pairs derived from 1,084 expert-curated images. |
| Approach: | They propose to use 4,264 question-answering (QA) pairs derived from 1,084 expert-curated images to evaluate the visual perception of multimodal large language models (MLLMs) . they also propose to utilize cross-panel relation understanding to evaluate MLLM’s ability to decipher intricate cross-panel relations. |
| Outcome: | The proposed model is based on 4,264 question-answering pairs derived from 1,084 expert-curated images. |
Field Embedding: A Unified Grain-Based Framework for Word Representation (2021.naacl-main)
Copied to clipboard
| Challenge: | Current methods focus on learning word embeddings while linguistic information is discarded after the learning. |
| Approach: | They propose a framework field embedding to jointly learn word and grain embedds by incorporating morphological, phonetic, and syntactical linguistic fields. |
| Outcome: | The proposed framework integrates morphological, phonetic, and syntactical linguistic fields to learn word embeddings and grain embedds. |
Sketch-Driven Regular Expression Generation from Natural Language and Examples (2020.tacl-1)
Copied to clipboard
| Challenge: | Recent systems for converting natural language descriptions into regexes have achieved some success, but typically deal with short, formulaic text and can only produce simple regexe. |
| Approach: | They propose a framework for regex synthesis in a context where both natural language and examples are available. |
| Outcome: | The proposed framework achieves state-of-the-art on two prior datasets and a real-world dataset, which existing neural systems completely fail on. |
Attention-Based Capsule Networks with Dynamic Routing for Relation Extraction (D18-1)
Copied to clipboard
| Challenge: | Existing neural networks focus on instance representation, and subsampling fails to retain precise spatial relationships between higher-level parts. |
| Approach: | They propose a neural approach based on capsule networks with attention mechanisms to extract relational information from a capsule. |
| Outcome: | The proposed method improves the precision of the predicted relations with different benchmarks. |
Sheaf Discovery with Joint Computation Graph Pruning and Flexible Granularity (2025.emnlp-main)
Copied to clipboard
| Challenge: | Experimental results show that DiscoGP extracts sheaves that preserve 93-100% of a model’s performance while comprising only 1-7% of the original weights and connections. |
| Approach: | They propose a framework for extracting self-contained modular units within neural language models (LMs) they use a gradient-based pruning algorithm to prune the original LM to a sparse skeleton . |
| Outcome: | The proposed framework preserves 93-100% of the original model's performance while preserving only 1-7% of the model''s original weights and connections. |
Feedback-Driven Tool-Use Improvements in Large Language Models via Automated Build Environments (2026.findings-acl)
Copied to clipboard
Junjie Ye, Changhao Jiang, Zhengyin Du, Yufei Xu, Xuesong Yao, Zhiheng Xi, Xiaoran Fan, Qi Zhang, Tao Gui, Xuanjing Huang, Jiecao Chen
| Challenge: | Currently, there are no efficient reinforcement learning (RL) frameworks specifically designed for tool use. |
| Approach: | They propose an automated environment construction pipeline that incorporates scenario decomposition, document generation, function integration, complexity scaling, and localized deployment to enable high-quality training environments without external tools. |
| Outcome: | The proposed framework significantly improves the models’ tool-use performance without degrading their general capabilities. |
CritiQ: Mining Data Quality Criteria from Human Preferences (2025.acl-long)
Copied to clipboard
Honglin Guo, Kai Lv, Qipeng Guo, Tianyi Liang, Zhiheng Xi, Demin Song, Qiuyinzhe Zhang, Yu Sun, Kai Chen, Xipeng Qiu, Tao Gui
| Challenge: | Existing methods to train language models rely on manual design, perplexity, or careful prompt engineering. |
| Approach: | They propose a method that automatically mines criteria from human preferences for data quality with only 30 human-annotated pairs and performs efficient data selection. |
| Outcome: | The proposed method improves on human-annotated test sets and shows high accuracy on code, math, and logic domains. |
Dialect-SQL: An Adaptive Framework for Bridging the Dialect Gap in Text-to-SQL (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing Text-to-SQL research focuses on specific database systems, limiting adaptability to different dialects. |
| Approach: | They propose a framework that employs Object Relational Mapping (ORM) code as an intermediate language to bridge this gap. |
| Outcome: | The proposed framework outperforms existing methods that generate SQL queries directly. |
Self-supervised Preference Optimization: Enhance Your Language Model with Preference Degree Awareness (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies have focused on replacing the reward model in Reinforcement Learning with Human Feedback (RLHF) methods for Large Language Models (LLMs). |
| Approach: | They propose a self-supervised preference optimization framework that replaces the reward model with a preference loss and alignment loss to improve LLMs' ability to understand human preferences. |
| Outcome: | The proposed framework can be integrated with existing preference optimization methods and significantly boost their performance. |
Detecting Customer Complaint Escalation with Recurrent Neural Networks and Manually-Engineered Features (N19-2)
Copied to clipboard
Wei Yang, Luchen Tan, Chunwei Lu, Anqi Cui, Han Li, Xi Chen, Kun Xiong, Muzi Wang, Ming Li, Jian Pei, Jimmy Lin
| Challenge: | e-commerce companies often have the option of escalating complaints by filing grievances with a government authority . this is detrimental to an ecommerce company, but this problem is challenging to solve by integrating recurrent neural networks with manually-engineered features. |
| Approach: | They propose a model that integrates recurrent neural networks with manually-engineered features to identify cases where the customer expresses such an intent. |
| Outcome: | The proposed model outperforms baseline models and provides better recall and triage for specialized agents. |
Unraveling the Mystery of Artifacts in Machine Generated Text (2022.lrec-1)
Copied to clipboard
| Challenge: | Recent studies show that human-written text is not distinguishable from synthetic text because of semantic errors or logical contradictions. |
| Approach: | They propose to analyze the forms of artifacts left by neural Text Generation Models by corrupting texts and replacing them with linguistic or statistical features. |
| Outcome: | The proposed method replaces text with linguistic or statistical features and improves the accuracy of the model. |
HopRAG: Multi-Hop Reasoning for Logic-Aware Retrieval-Augmented Generation (2025.findings-acl)
Copied to clipboard
| Challenge: | Traditional retrieval systems focus on lexical or semantic similarity rather than logical relevance. |
| Approach: | They propose a new RAG framework that augments retrieval with logical reasoning . hopRAG uses a retrieve-reason-prune mechanism to explore multi-hop neighbors . |
| Outcome: | The proposed framework outperforms conventional retrieval systems and state-of-the-art benchmarks on multi-hop QA tasks. |
ToolHop: A Query-Driven Benchmark for Evaluating Large Language Models in Multi-Hop Tool Use (2025.acl-long)
Copied to clipboard
Junjie Ye, Zhengyin Du, Xuesong Yao, Weijian Lin, Yufei Xu, Zehui Chen, Zaiyuan Wang, Sining Zhu, Zhiheng Xi, Siyu Yuan, Tao Gui, Qi Zhang, Xuanjing Huang, Jiecao Chen
| Challenge: | Effective evaluation of multi-hop tool use is critical for analyzing the understanding, reasoning, and function-calling capabilities of large language models. |
| Approach: | They propose a dataset that provides rigorous evaluation of multi-hop tool use. |
| Outcome: | The proposed model achieves 49.04% accuracy across five model families. |
SynWorld: Virtual Scenario Synthesis for Agentic Action Knowledge Refinement (2025.acl-short)
Copied to clipboard
Runnan Fang, Xiaobin Wang, Yuan Liang, Shuofei Qiao, Jialong Wu, Zekun Xi, Ningyu Zhang, Yong Jiang, Pengjun Xie, Fei Huang, Huajun Chen
| Challenge: | Using Large Language Models (LLMs)-based agents can enhance their understanding of environments and tasks. |
| Approach: | They propose a framework that allows agents to synthesize possible scenarios with multi-step action invocation within the action space and perform Monte Carlo Tree Search exploration to refine their action knowledge in the current environment. |
| Outcome: | The proposed framework synthesizes possible scenarios with multi-step action invocation within the action space and performs Monte Carlo Tree Search exploration to refine action knowledge in the current environment. |
Into the Gray Zone: Domain Contexts Can Blur LLM Safety Boundaries (2026.acl-long)
Copied to clipboard
Ki Sen Hung, Xi Yang, Chang Liu, Haoran Li, Kejiang Chen, Changxuan Fan, Tsun On Kwok, Weiming Zhang, Xiaomeng Li, Yangqiu Song
| Challenge: | a goal of LLM alignment is to balance usefulness with harmlessness, but this conflictes when knowledge serves both legitimate and malicious purposes. |
| Approach: | They propose a framework that combines safety-research contexts with adversarial interactions to exploit a vulnerability in Jargon queries. |
| Outcome: | a framework outperforms existing methods in analyzing Jargon queries, a study shows . it achieves 93% of attacks across seven models, while remaining useful, the authors say . |
M2Edit: Locate and Edit Multi-Granularity Knowledge in Multimodal Large Language Model (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing knowledge editing methods for MLLMs lack multi-granularity knowledge . existing knowledge editing approaches lack multimodality knowledge and generalize to multimodal data. |
| Approach: | They propose a multimodal knowledge editing method which integrates key knowledge layers within MLLMs and collaboratively edits them. |
| Outcome: | The proposed method improves visual generality performance on knowledge data of different granularities. |
SampleMix: A Sample-wise Pre-training Data Mixing Strategy by Coordinating Data Quality and Diversity (2025.findings-emnlp)
Copied to clipboard
Xiangyu Xi, Deyang Kong, Jian Yang, Jiawei Yang, Zhengyu Chen, Wei Wang, Jingang Wang, Xunliang Cai, Shikun Zhang, Wei Ye
| Challenge: | Existing methods for pretraining data mixing for large language models neglect significant inter-domain overlaps and commonalities, failing to control the global diversity of the constructed training dataset. |
| Approach: | They propose a sample-wise data mixture approach that performs global cross-domain sampling by systematically evaluating the quality and diversity of each sample. |
| Outcome: | The proposed method exceeds existing domain-based methods in multiple downstream tasks and perplexity assessments. |
Think Earlier, Not Longer: Prompt Optimization via Reducing Unhealthy Exploration (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing approaches to improve reasoning performance ignore the presence of unhealthy exploration that increases token usage without contributing to effective problem-solving. |
| Approach: | They propose an entropy-dynamics-aware prompt optimization framework that trains a lightweight optimizer to generate concise clarifications. |
| Outcome: | The proposed framework reduces ambiguity-induced early-stage uncertainty while preserving the model's reasoning capabilities. |
HoLLMwood: Unleashing the Creativity of Large Language Models in Screenwriting via Role Playing (2024.findings-emnlp)
Copied to clipboard
Jing Chen, Xinyu Zhu, Cheng Yang, Chufan Shi, Yadong Xi, Yuxiang Zhang, Junjie Wang, Jiashu Pu, Tian Feng, Yujiu Yang, Rongsheng Zhang
| Challenge: | Generative AI has demonstrated unprecedented creativity in the field of computer vision, yet such phenomena have not been observed in the realm of literary creation. |
| Approach: | They propose a framework for unleashing the creativity of large language models (LLMs) they assign LLMs to different roles involved in real-world scenario, they write . |
| Outcome: | The proposed framework outperforms baselines in terms of coherence, relevance, interestingness and overall quality on automatically generated screenplays. |
Beyond Single-shot Writing: Deep Research Agents are Unreliable at Multi-turn Report Revision (2026.acl-long)
Copied to clipboard
| Challenge: | Existing benchmarks for Deep Research Agents (DRAs) treat report generation as a single-shot writing task. |
| Approach: | They propose an evaluation suite that establishes multi-turn report revision as a new axis. |
| Outcome: | The evaluation suite establishes multi-turn report revision as a new axis. |
Context-Driven Index Trimming: A Data Quality Perspective to Enhancing Precision of RALMs (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing research often overlooks the data quality issues within retrieval results, often caused by inaccurate existing vector-distance-based retrieval methods. |
| Approach: | They propose to use Context-Driven Index Trimming framework to capture and regulate consistency between retrieved contexts and modify indexes in the database. |
| Outcome: | Experiments show that the proposed framework can improve answer quality by 3.75% on open-domain question-answering tasks. |
KuiLeiXi: a Chinese Open-Ended Text Adventure Game (2021.acl-demo)
Copied to clipboard
Yadong Xi, Xiaoxi Mao, Le Li, Lei Lin, Yanjiang Chen, Shuhan Yang, Xuhan Chen, Kailun Tao, Zhi Li, Gongzheng Li, Lin Jiang, Siyan Liu, Zeng Zhao, Minlie Huang, Changjie Fan, Zhipeng Hu
| Challenge: | Recent advances in pre-trained language models have made it possible to generate human-like text. |
| Approach: | They propose to integrate an open-ended text adventure game in Chinese, named KuiLeiXi, where players interact with the AI until the plot goals are reached. |
| Outcome: | The proposed game lacks incentives and relies on players to explore on their own. |