Papers by Zheng Cheng

Enhancing LLM-as-a-Judge through Active-Sampling-based Prompt Optimization (2025.acl-industry)

Challenge:	generative models struggle with logic-intensive instruction following, exposing a persistent reasoning–execution gap.
Approach:	They propose a task-agnostic reasoning architecture for general image generation . they propose pixel-level feedback to ground the Thinker's policy in pixel feedback .
Outcome:	The proposed system significantly improves image reasoning and generation quality.

Copied to clipboard

Cheng Zhen, Ervine Zheng, Jilong Kuang, Geoffrey Jay Tso

Low-Resource NLP

Challenge:	Suboptimal prompts can introduce biases, inconsistencies, and unreliable evaluations.
Approach:	They propose an active-sampling-based framework for automatic prompt optimization . they use a small, diverse subset of samples to guide prompt refinement .
Outcome:	The proposed framework outperforms baselines on four popular LLMs and three real-world datasets.

Towards Knowledge Checking in Retrieval-augmented Generation: A Representation Perspective (2025.naacl-long)

Copied to clipboard

Shenglai Zeng, Jiankun Zhang, Bingheng Li, Yuping Lin, Tianqi Zheng, Dante Everaert, Hanqing Lu, Hui Liu, Hui Liu, Yue Xing, Monica Xiao Cheng, Jiliang Tang

StoryAnalogy: Deriving Story-level Analogies from Large Language Models to Unlock Analogical Understanding (2023.emnlp-main)

Challenge:	Existing studies have shown that LLMs struggle to identify the boundaries of their own knowledge and tend to prioritize external information over internal knowledge learned during pre-training.
Approach:	They conduct a comprehensive analysis of LLM representation behaviors and demonstrate the significance of using representations in knowledge checking.
Outcome:	The proposed classifiers improve performance even when dealing with noisy knowledge databases.

Copied to clipboard

Cheng Jiayang, Lin Qiu, Tsz Chan, Tianqing Fang, Weiqi Wang, Chunkit Chan, Dongyu Ru, Qipeng Guo, Hongming Zhang, Yangqiu Song, Yue Zhang, Zheng Zhang

Attn-GS: Attention-Guided Context Compression for Efficient Personalized LLMs (2026.acl-long)

Challenge:	Analogy-making between narratives is crucial for human reasoning . despite its importance, there has been limited research on story analogies .
Approach:	They construct a large-scale story-level analogy corpus with 24K story pairs . they find that the tasks are incredibly difficult for large language models such as ChatGPT .
Outcome:	The proposed corpus contains 24K story pairs from diverse domains with human annotations on two similarities from the extended Structure-Mapping Theory.

Copied to clipboard

Shenglai Zeng, Tianqi Zheng, Chuan Tian, Dante Everaert, Yau-Shian Wang, Yupin Huang, Michael J. Morais, Rohit Patki, Jinjin Tian, Xinnan Dai, Kai Guo, Monica Xiao Cheng, Hui Liu

Green & Sustainable NLP

Challenge:	Existing approaches to personalize large language models (LLMs) rely on heuristic methods to compress user profiles but they ignore how LLMs process and prioritize different profile components.
Approach:	They propose an attention-guided context compression framework that leverages attention feedback from a marking model to mark important personalization sentences and guides a compression model to generate task-relevant compressed user contexts.
Outcome:	The proposed framework outperforms baselines across tasks, token limits, and settings while reducing token usage by 50 times.

Data Interpreter: An LLM Agent for Data Science (2025.findings-acl)

Copied to clipboard

Sirui Hong, Yizhang Lin, Bang Liu, Bangbang Liu, Binhao Wu, Ceyao Zhang, Danyang Li, Jiaqi Chen, Jiayi Zhang, Jinlin Wang, Li Zhang, Lingyao Zhang, Min Yang, Mingchen Zhuge, Taicheng Guo, Tuo Zhou, Wei Tao, Robert Tang, Xiangtao Lu, Xiawu Zheng, Xinbing Liang, Yaying Fei, Yuheng Cheng, Yongxin Ni, Zhibin Gou, Zongze Xu, Yuyu Luo, Chenglin Wu

Hephaestus: Improving Fundamental Agent Capabilities of Large Language Models through Continual Pre-Training (2025.naacl-long)

Challenge:	Large Language Models (LLMs) excel in various domains but face challenges when applied to data science workflows due to their complex, multi-stage nature.
Approach:	They propose a hierarchical graph-based agent that represents complexity and a progressive strategy for step-by-step verification, refinement, and consistent context management.
Outcome:	The proposed agent surpasses state-of-the-art baselines on the MATH dataset and performs better on InfiAgent-DABench.

Copied to clipboard

Yuchen Zhuang, Jingfeng Yang, Haoming Jiang, Xin Liu, Kewei Cheng, Sanket Lokegaonkar, Yifan Gao, Qing Ping, Tianyi Liu, Binxuan Huang, Zheng Li, Zhengyang Wang, Pei Chen, Ruijie Wang, Rongzhi Zhang, Nasser Zalmout, Priyanka Nigam, Bing Yin, Chao Zhang

Enhancing Uncertainty-Based Hallucination Detection with Stronger Focus (2023.emnlp-main)

Challenge:	Existing LLMs often rely on complex prompting or extensive fine-tuning to introduce new capabilities while preserving strong generalizability.
Approach:	They propose a large-scale pre-training corpus to enhance LLM agents' capabilities . they use 103B agent-specific data encompassing 76,537 APIs .
Outcome:	The proposed training corpus outperforms open-source LLMs and commercial LLM agents on three agent benchmarks.

Copied to clipboard

Tianhang Zhang, Lin Qiu, Qipeng Guo, Cheng Deng, Yue Zhang, Zheng Zhang, Chenghu Zhou, Xinbing Wang, Luoyi Fu

PaCoRe: Learning to Scale Test-Time Compute with Parallel Coordinated Reasoning (2026.acl-long)

Challenge:	Existing methods for detecting hallucinations in LLMs rely on external knowledge for reference retrieval or require sampling multiple responses for consistency verification.
Approach:	They propose a reference-free, uncertainty-based method for detecting hallucinations in Large Language Models that imitates human focus in factuality checking from three aspects: focus on the most informative keywords; focus on unreliable tokens in historical context; focus of token properties such as token type and token frequency.
Outcome:	The proposed method achieves state-of-the-art performance across all evaluation metrics and eliminates the need for additional information.

Copied to clipboard

Jingcheng Hu, Yinmin Zhang, Shijie Shang, Xiaobo Yang, Yue Peng, Zhewei Huang, Hebin Zhou, Xin Wu, Jie Cheng, Fanqi Wan, Xiangwen Kong, Chengyuan Yao, Kaiwen Yan, Ailin Huang, Hongyu Zhou, Qi Han, Zheng Ge, Xiangyu Zhang, Heung-Yeung Shum

PERM: Psychology-grounded Empathetic Reward Modeling for Large Language Models (2026.findings-acl)

Challenge:	Parallel Coordinated Reasoning (PaCoRe) overcomes a central limitation of contemporary language models: their inability to scale test-time compute (TTC) far beyond sequential reasoning under a fixed context window.
Approach:	They propose a training-and-inference framework to overcome a central limitation of language models: their inability to scale test-time compute (TTC) under a fixed context window.
Outcome:	The proposed model scales to multi-million-token effective TTC without exceeding context limits.

Copied to clipboard

Chengbing Wang, Wuqiang Zheng, Yang Zhang, Fengbin Zhu, Junyi Cheng, Yi Xie, Wenjie Wang, Fuli Feng

Guiding the Growth: Difficulty-Controllable Question Generation through Step-by-Step Rewriting (2021.acl-long)

Challenge:	Existing reward models evaluate empathy from a single perspective, overlooking bidirectional interaction nature of empathy.
Approach:	They propose a reward model that evaluates empathy from a single perspective . they propose PERM to integrate a bystander perspective to monitor overall interaction quality .
Outcome:	a new reward model outperforms state-of-the-art models on an emotional intelligence benchmark and an industrial daily conversation dataset.

Copied to clipboard

Yi Cheng, Siyao Li, Bang Liu, Ruihui Zhao, Sujian Li, Chenghua Lin, Yefeng Zheng

Text Generation

Challenge:	Existing QG systems perform substantially worse in answering multi-hop questions than single-hop ones.
Approach:	They propose a framework that progressively increases question difficulty through step-by-step rewriting under the guidance of an extracted reasoning chain.
Outcome:	The proposed framework increases question difficulty through step-by-step rewriting under the guidance of an extracted reasoning chain.

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.

Flooding-X: Improving BERT’s Resistance to Adversarial Attacks via Loss-Restricted Fine-Tuning (2022.acl-long)

Copied to clipboard

Qin Liu, Rui Zheng, Bao Rong, Jingyi Liu, ZhiHua Liu, Zhanzhan Cheng, Liang Qiao, Tao Gui, Qi Zhang, Xuanjing Huang

Robustness in NLP

Challenge:	Existing approaches to generating adversarial perturbations scale up the cost of training computational complexity by the number of gradient steps it takes to obtain the adversarials.
Approach:	They propose a flood method which aims at better generalization and a criterion to bring hyper-parameter-dependent flooding into effect with a narrowed-down search space by measuring how the gradient steps taken within one epoch affect the loss of each batch.
Outcome:	The proposed method improves BERT’s resistance to textual adversarial attacks by a large margin and achieves state-of-the-art robust accuracy on various text classification and GLUE tasks.

From Implicit Graph Encoding to Explicit Evidence: A Training-Free LLM Framework for Temporal Knowledge Graph Reasoning (2026.findings-acl)

Copied to clipboard

Guo Tang, Ke Cheng, Huiming Fan, Heng Chang, Wenxiang Zheng, Xianhao Ou, Junjia Xiang, Ming Liu, Yujun Zhou, Li Lanyu, Bing Qin

Evaluating Memory Capability in Continuous Lifelog Scenario (2026.findings-acl)

Challenge:	Existing Large Language Models (LLMs) struggle with implicit modality alignment and suboptimal graph linearization.
Approach:	They propose a training-free, test-time adaptive framework that reframes TKG prediction as explicit evidence-driven reasoning.
Outcome:	ExE-LLM outperforms fully trained graph neural networks on four benchmarks . it achieves SOTA performance in inductive settings, significantly outperforming fully trained neural networks .

Copied to clipboard

Jianjie Zheng, Zhichen Liu, Zhanyu Shen, Jingxiang Qu, Guanhua Chen, Yile Wang, Yang Xu, Yang Liu, Sijie Cheng

Challenge:	Existing benchmarks focus on online one-on-one chatting or human-AI interactions, neglecting real-world scenarios.
Approach:	They propose a framework to curate a lifelog benchmark that combines two subsets of audio data to address temporal leakage in offline settings.
Outcome:	The proposed framework outperforms existing benchmarks on live chats and AI interactions.

DualRAG: A Dual-Process Approach to Integrate Reasoning and Retrieval for Multi-Hop Question Answering (2025.acl-long)

Copied to clipboard

Rong Cheng, Jinyi Liu, Yan Zheng, Fei Ni, Jiazhen Du, Hangyu Mao, Fuzheng Zhang, Bo Wang, Jianye Hao

Natural Language Interfaces

Challenge:	Existing approaches to multi-hop question answering struggle to identify and organize dynamic knowledge . et al., 2023; Liu e.t. al. 2023) suggest a dual-process framework for multi-step reasoning .
Approach:	They propose a synergistic dual-process framework that integrates reasoning and retrieval.
Outcome:	The proposed framework improves answer accuracy and coherence even in smaller-scale models.

Evaluating and Enhancing the Robustness of Neural Network-based Dependency Parsing Models with Adversarial Examples (2020.acl-main)

Copied to clipboard

Xiaoqing Zheng, Jiehang Zeng, Yi Zhou, Cho-Jui Hsieh, Minhao Cheng, Xuanjing Huang

Robustness in NLP

Challenge:	Previously studies focused on semantic tasks such as sentiment analysis, question answering and reading comprehension.
Approach:	They propose two approaches to study where and how adversarial examples exist in dependency parsing . they use a state-of-the-art parser to find adversarials in existing texts .
Outcome:	The proposed approaches show that adversarial examples exist in dependency parsing . they show that up to 77% of input examples admit adversarials .

MoRI: Learning Motivation-Grounded Reasoning for Scientific Ideation in Large Language Models (2026.acl-long)

Copied to clipboard

Chenyang Gu, Jiahao Cheng, Meicong Zhang, Pujun Zheng, Jinquan Zheng, Guoxiu He

DP3: Differentially Private Prompt Perturbation for Multi-turn LLM Inference (2026.findings-acl)

Challenge:	Existing LLMs emulate human research workflows but lack scientific grounding . empirical results show that MoRI outperforms strong commercial LLM models .
Approach:	They propose a framework that explicitly learns scientific reasoning from research motivations to methodologies.
Outcome:	The proposed framework outperforms commercial LLMs and agentic baselines in novelty, technical rigor, and feasibility.

Copied to clipboard

Lele Zheng, Chao Zhang, Feiyang Yuan, Ke Cheng, Tao Zhang, Anxiao Song, Yulong Shen

Empirical Study on Data Attributes Insufficiency of Evaluation Benchmarks for LLMs (2025.coling-main)

Challenge:	Large language models (LLMs) are widely used for text understanding and generation . existing methods that assume single-turn interactions break down in multi-turn settings .
Approach:	They propose a differentially private prompt perturbation framework for multi-turn LLM inference . DP3 constructs a perturbation mapping table to reuse perturbations for recurring tokens .
Outcome:	The proposed framework reduces privacy costs and degrades cross-turn semantic coherence . it also provides a context-aware utility function to maintain semantic consistency across turns .

Copied to clipboard

Chuang Liu, Renren Jin, Zheng Yao, Tianyi Li, Liang Cheng, Mark Steedman, Deyi Xiong

Challenge:	Existing benchmarks for evaluating large language models neglect key qualitative data attributes that can significantly impact the final rankings of LLMs.
Approach:	They propose a framework with three modules designed to assess diversity, redundancy, and difficulty.
Outcome:	The proposed framework systematically incorporates diversity, redundancy, and difficulty attributes and shows that they influence the ranking of LLMs.

OS-Kairos: Adaptive Interaction for MLLM-Powered GUI Agents (2025.findings-acl)

Copied to clipboard

Pengzhou Cheng, Zheng Wu, Zongru Wu, Tianjie Ju, Aston Zhang, Zhuosheng Zhang, Gongshen Liu

Green & Sustainable NLP

Challenge:	Existing efforts to build GUI agents focused on the autonomous mode have failed to address the problem of over-execution.
Approach:	They propose an adaptive GUI agent that predicts confidence levels at each interaction step and elicits adaptive interaction.
Outcome:	The proposed GUI agent outperforms existing models on a complex dataset and on established benchmarks.

Reinforcement Learning on Pre-Training Data (2026.acl-long)

Copied to clipboard

Siheng Li, Kejiao Li, Zenan Xu, Guanhua Huang, Kun Li, Haoyuan Wu, null Wujiajia, Zihao Zheng, Chenchen Zhang, Kun Shi, Xue Gong, Qi Yi, Ruibin Xiong, Tingqiang Xu, Yuhao Jiang, Jianfeng Yan, Yuyuan Zeng, Guanghui Xu, Jinbao Xue, Zhijiang xu, Zheng Fang, Shuai LI, Qibin Liu, Xiaoxue Li, Zhuoyu Li, Yangyu Tao, Fei Gao, Cheng Jiang, Bochao Wang, Kai Liu, Jianchen Zhu, Wai Lam, Bo Zhou, Di Wang

Improving Multi-turn Emotional Support Dialogue Generation with Lookahead Strategy Planning (2022.emnlp-main)

Challenge:	Recent progress in large language models is driven by scaling of training compute through pre-training with nexttoken prediction (NTP) or post-training (RL) Pre-training using NTP enables models to acquire extensive knowledge and skills from general data, but it suffers from data inefficiency and catastrophic forgetting in continual learning settings.
Approach:	They propose to scale training compute through pre-training with next-token prediction (NTP) or post-training by scaling reinforcement learning (RL) to improve learning from general data.
Outcome:	Experiments on multiple benchmarks and models show that the proposed approach improves continual pre-training and provides a strong foundation for post-training on Qwen3-8B-Base.

Copied to clipboard

Yi Cheng, Wenge Liu, Wenjie Li, Jiashuo Wang, Ruihui Zhao, Bang Liu, Xiaodan Liang, Yefeng Zheng

Dialogue Systems & Conversational Agents

Challenge:	Existing research on building ES conversation systems only considered single-turn interactions with users, which is over-simplified and has limited support for multi-turn systems.
Approach:	They propose a multi-turn ES conversation system that uses lookahead heuristics to estimate future user feedback after using particular strategies.
Outcome:	The proposed system significantly outperforms baselines in both dialogue generation and strategy planning.

MovieChats: Chat like Humans in a Closed Domain (2020.emnlp-main)

Copied to clipboard

Hui Su, Xiaoyu Shen, Zhou Xiao, Zheng Zhang, Ernie Chang, Cheng Zhang, Cheng Niu, Jie Zhou

Natural Language Interfaces

Challenge:	Currently, open-domain chatbots are far from satisfactory.
Approach:	They propose a unified, readily scalable neural approach which reconciles all subtasks like intent prediction and knowledge retrieval.
Outcome:	The proposed approach outperforms commercial systems replying on complex rules on static and interactive tests and shows that the results are remarkably good.

On the Safety of Conversational Models: Taxonomy, Dataset, and Benchmark (2022.findings-acl)

Copied to clipboard

Hao Sun, Guangxuan Xu, Jiawen Deng, Jiale Cheng, Chujie Zheng, Hao Zhou, Nanyun Peng, Xiaoyan Zhu, Minlie Huang

Dialogue Systems & Conversational Agents

Challenge:	Dialogue safety problems severely limit the real-world deployment of generative conversational models.
Approach:	They propose a taxonomy for dialogue safety specifically designed to capture unsafe behaviors in human-bot dialogue settings.
Outcome:	The proposed taxonomy captures unsafe behaviors in human-bot dialogue settings with rich context-sensitive unsafe examples.

Black-Box Prompt Optimization: Aligning Large Language Models without Model Training (2024.acl-long)

Copied to clipboard

Jiale Cheng, Xiao Liu, Kehan Zheng, Pei Ke, Hongning Wang, Yuxiao Dong, Jie Tang, Minlie Huang

LogiDynamics: Unraveling the Dynamics of Inductive, Abductive and Deductive Logical Inferences in LLM Reasoning (2025.emnlp-main)

Challenge:	Large language models are often not well aligned with human intents, which requires additional training.
Approach:	They propose to use Black-Box Prompt Optimization (BPO) to perform alignments on large language models that are not well aligned with human intents.
Outcome:	The proposed model outperforms existing models and is model-agnostic.

Copied to clipboard

Tianshi Zheng, Cheng Jiayang, Chunyang Li, Haochen Shi, Zihao Wang, Jiaxin Bai, Yangqiu Song, Ginny Wong, Simon See

OmniFlatten: An End-to-end GPT Model for Seamless Voice Conversation (2025.acl-long)

Challenge:	Modern large language models (LLMs) employ diverse logical inference mechanisms for reasoning.
Approach:	They analyze the comparative dynamics of inductive (System 1) versus abductive/deductive (system 2) inference in large language models by using a controlled analogical reasoning environment and a MCQ/free-text task format.
Outcome:	The proposed methods can significantly scale LLM reasoning.

Copied to clipboard

Qinglin Zhang, Luyao Cheng, Chong Deng, Qian Chen, Wen Wang, Siqi Zheng, Jiaqing Liu, Hai Yu, Chao-Hong Tan, Zhihao Du, ShiLiang Zhang

Dialogue Systems & Conversational Agents

Challenge:	Full-duplex spoken dialogue systems allow simultaneous bidirectional communication . low latency and natural interactions in full-duplice systems remains a challenge .
Approach:	They propose a multi-stage post-training scheme that adapts a text large language model into a speech-text dialogue LLM.
Outcome:	The proposed model can model human conversation behaviors with low latency and natural interactions with low delay.

Controlling Styles in Neural Machine Translation with Activation Prompt (2023.findings-acl)

Copied to clipboard

Yifan Wang, Zewei Sun, Shanbo Cheng, Weiguo Zheng, Mingxuan Wang

Machine Translation

Challenge:	Earlier studies on controlling styles in neural machine translation (NMT) have focused on regulating the level of formality, but they still encounter two major challenges.
Approach:	They propose a method to control the style of neural machine translation by retrieving prompts from stylized monolingual corpus.
Outcome:	The proposed method can control the style of translation and achieve remarkable performance.

CapArena: Benchmarking and Analyzing Detailed Image Captioning in the LLM Era (2025.findings-acl)

Copied to clipboard

Kanzhi Cheng, Wenpo Song, Jiaxin Fan, Zheng Ma, Qiushi Sun, Fangzhi Xu, Chenyang Yan, Nuo Chen, Jianbing Zhang, Jiajun Chen

G-Cap: A Game Character Caption Generator (2026.acl-long)

Challenge:	Image captioning has been a challenge for vision-language researchers for decades . current VLMs focus on tasks like visual question answering (YA) but image captioning is not as advanced as expected.
Approach:	They evaluate VLMs' performance on image captioning using human annotations . they find that some metrics show high caption-level agreement with humans .
Outcome:	The proposed model outperforms open-source models on image captioning . it achieves 93.4% correlation with human rankings at $4 per test .

Copied to clipboard

Yang Yang, Feng Hu, Haiming Zhang, XU Cheng, Gui Zheng, Liang Yao, Wenqi Ren

Text Generation

Challenge:	Existing studies on Large Vision-Language Models (LVLMs) primarily focus on real-world scenarios, leaving surreal, highly stylized, and semantically hybrid virtual-world situations significantly underexplored.
Approach:	They propose to use a manually annotated benchmark to evaluate LVLMs' ability to perceive and describe game character from the virtual-world.
Outcome:	The proposed task evaluates LVLMs’ ability to perceive and describe game character from the virtual-world.

Integrating Audio, Visual, and Semantic Information for Enhanced Multimodal Speaker Diarization on Multi-party Conversation (2025.acl-long)

Copied to clipboard

Luyao Cheng, Hui Wang, Chong Deng, Siqi Zheng, Yafeng Chen, Rongjie Huang, Qinglin Zhang, Qian Chen, Xihao Li, Wen Wang

Hidden Ghost Hand: Unveiling Backdoor Vulnerabilities in MLLM-Powered Mobile GUI Agents (2025.findings-emnlp)

Challenge:	Mainstream speaker diarization systems rely only on acoustic information, making it challenging in complex aural environments.
Approach:	They propose a multimodal approach that integrates audio, visual, and semantic cues to enhance speaker diarization.
Outcome:	The proposed approach outperforms state-of-the-art methods on multi-party conversations . it integrates audio-visual-semantic cues into the clustering process for acoustic speaker embeddings .

Copied to clipboard

Pengzhou Cheng, Haowen Hu, Zheng Wu, Zongru Wu, Tianjie Ju, Daizong Ding, Zhuosheng Zhang, Gongshen Liu

Robustness in NLP

Challenge:	MLLM-powered GUI agents expose multiple interaction-level triggers, causing backdoor attacks . backdoor injection maximizes feature difference across sample classes, improving flexibility .
Approach:	They propose a framework for red-teaming backdoor attacks using MLLMs . they construct composite triggers by combining goal and interaction levels .
Outcome:	The proposed framework is effective and stealthy for red-teaming backdoor attacks.

Learning to Evolve: A Self-Improving Framework for Multi-Agent Systems via Textual Parameter Graph Optimization (2026.findings-acl)

Copied to clipboard

Shan He, Runze Wang, Zhuoyun Du, Huiyu Bai, Zouying Cao, Yu Cheng, Bo Zheng

Structured Data in NLP

Challenge:	Existing methods for designing and optimizing multi-agent systems are static and do not learn from experience.
Approach:	They propose a framework that enables a multi-agent system to learn to evolve . they use "textual gradients" to pinpoint failures and suggest granular modifications .
Outcome:	a new framework enables a multi-agent system to learn to evolve . it learns from historical optimization experiences to improve its performance .

OpenResearcher: Unleashing AI for Accelerated Scientific Research (2024.emnlp-demo)

Copied to clipboard

Yuxiang Zheng, Shichao Sun, Lin Qiu, Dongyu Ru, Cheng Jiayang, Xuefeng Li, Jifan Lin, Binjie Wang, Yun Luo, Renjie Pan, Yang Xu, Qingkai Min, Zizhao Zhang, Yiwen Wang, Wenjie Li, Pengfei Liu

Information Retrieval

Challenge:	Global scientific publications are growing annually by about 4%-5% (Pinedo et al., 2024).
Approach:	They introduce an AI-assisted platform that answers diverse questions from researchers using Retrieval-Augmented Generation (RAG) they develop various tools to understand queries, search from the scientific literature, filter retrieved information, provide accurate and comprehensive answers, and self-refine answers.
Outcome:	OpenResearcher is built on Retrieval-Augmented Generation (RAG) to integrate Large Language Models (LLMs) with up-to-date, domain-specific knowledge.

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

Perceive the Passage of Time: A Systematic Evaluation of Large Language Model in Temporal Relativity (2025.coling-main)

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.

Copied to clipboard

Shuang Chen, Yining Zheng, Shimin Li, Qinyuan Cheng, Xipeng Qiu

R3-RAG: Learning Step-by-Step Reasoning and Retrieval for LLMs via Reinforcement Learning (2025.findings-emnlp)

Challenge:	Temporal perception is crucial for Large Language Models to understand the world.
Approach:	They propose a temporal-relative ability benchmark to evaluate LLMs' temporal perception . they conduct extensive experiments on popular LLM GPT-4 scenarios .
Outcome:	The proposed benchmarks show a significant performance gap between LLMs and humans in temporal-relative capability.

Copied to clipboard

Yuan Li, Qi Luo, Xiaonan Li, Bufan Li, Qinyuan Cheng, Bo Wang, Yining Zheng, Yuxin Wang, Zhangyue Yin, Xipeng Qiu

Enabling Agents to Communicate Entirely in Latent Space (2026.acl-long)

Challenge:	RAG systems that integrate external knowledge with Large Language Models often become bottlenecks due to their limited parameters compared to LLMs and their inability to perform step-by-step reasoning.
Approach:	They propose a model that integrates external knowledge with Large Language Models to enhance factual correctness and mitigate hallucination.
Outcome:	The proposed model outperforms baselines and can transfer well to different retrievers.

Copied to clipboard

Zhuoyun Du, Runze Wang, Huiyu Bai, Zouying Cao, Xiaoyong Zhu, Yu Cheng, Bo Zheng, Wei Chen, Haochao Ying

Challenge:	Natural language is the de facto communication medium for LLM-based agents, but it presents a fundamental constraint . natural language downsampling limits the depth and nuance of information that can be transmitted . et al.: inter-agent latent space communication is a promising paradigm for solving complex tasks .
Approach:	They propose a paradigm that leverages the last hidden states of an LLM as a representation of its thought for direct communication.
Outcome:	The proposed paradigm outperforms fine-tuned chain-of-thought prompting and single-agent baselines even across heterogeneous models.

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

Sample-Efficient Human Evaluation of Large Language Models via Maximum Discrepancy Competition (2025.acl-long)

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.

Copied to clipboard

Kehua Feng, Keyan Ding, Tan Hongzhi, Kede Ma, Zhihua Wang, Shuangquan Guo, Cheng Yuzhou, Ge Sun, Guozhou Zheng, Qiang Zhang, Huajun Chen

EventGround: Narrative Reasoning by Grounding to Eventuality-centric Knowledge Graphs (2024.lrec-main)

Challenge:	Existing methods for evaluation of large language models are inefficient and inefficient due to inaccuracy of standard metrics in human perception of text quality and inefficiency in sampling informative test examples.
Approach:	They propose a sample-efficient human evaluation method for large language models based on the principle of MAximum Discrepancy (MAD) competition.
Outcome:	The proposed method achieves the “golden” ranking of LLMs with a minimum set of input instructions, which in turn reveal their relative strengths and weaknesses.

Copied to clipboard

Cheng Jiayang, Lin Qiu, Chunkit Chan, Xin Liu, Yangqiu Song, Zheng Zhang

Finch: Benchmarking Finance & Accounting across Spreadsheet-Centric Enterprise Workflows (2026.findings-acl)

Challenge:	Existing frameworks for leveraging background knowledge of narratives are limited.
Approach:	They propose a framework to ground free-texts to eventuality-centric KGs for narrative reasoning . their framework is based on a set of probabilistic probabilistic models that are grounded in the real world .
Outcome:	The proposed framework outperforms baseline models while providing interpretable evidence.

Copied to clipboard

Haoyu Dong, Pengkun Zhang, Yan Gao, Xuanyu Dong, Yilin Cheng, Mingzhe Lu, Adina Yakefu, Shuxin Zheng

LLMBox: A Comprehensive Library for Large Language Models (2024.acl-demos)

Challenge:	FinWorkBench evaluates real-world enterprise-grade finance and accounting workflows . a human evaluation of GPT 5.1 Pro passes only 38.4% of workflows, a study finds .
Approach:	They propose a workflow construction process that combines LLM-assisted mining and expert annotation to build 172 composite workflows.
Outcome:	The proposed process combines expert annotation with LLM-assisted mining of workflows from authentic enterprise environments.

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

PRISM-MCTS: Learning from Reasoning Trajectories with Metacognitive Reflection (2026.findings-acl)

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.

Copied to clipboard

Siyuan Cheng, Bozhong Tian, Yanchao Hao, Zheng Wei

Alignment before Awareness: Towards Visual Question Localized-Answering in Robotic Surgery via Optimal Transport and Answer Semantics (2024.lrec-main)

Challenge:	Existing reasoning models are limited by inefficiency and computational redundancy . PRISM-MCTS integrates a process reward model with a dynamic shared memory .
Approach:	They propose a reasoning framework that integrates a process reward model with a dynamic shared memory.
Outcome:	PRISM-MCTS integrates a process reward model with a dynamic shared memory . it halves trajectory requirements on GPQA while surpassing MCTS-RAG and Search-o1 .

Copied to clipboard

Zhihong Zhu, Yunyan Zhang, Xuxin Cheng, Zhiqi Huang, Derong Xu, Xian Wu, Yefeng Zheng

DocHieNet: A Large and Diverse Dataset for Document Hierarchy Parsing (2024.emnlp-main)

Challenge:	Recent models for visual question localized-answering (VQLA) lack the ability to relate these answers to their localization at an instance level.
Approach:	They propose a model which introduces optimal transport to achieve bidirectional and fine-grained alignment between images and questions, enabling more precise localization.
Outcome:	The proposed model outperforms state-of-the-art models on two widely-used datasets on surgical scenes and surgical instruments.

Copied to clipboard

Hangdi Xing, Changxu Cheng, Feiyu Gao, Zirui Shao, Zhi Yu, Jiajun Bu, Qi Zheng, Cong Yao

Challenge:	Existing methods for document hierarchy parsing are limited due to the small scale and inconsistency of datasets.
Approach:	They propose a document hierarchy parsing dataset to compensate for the data scarcity problem and propose 'dHP' framework to grasp fine-grained text content and coarse-grounded pattern at layout element level.
Outcome:	The proposed framework grasps both fine-grained text content and coarse-grounded pattern at layout element level, enhancing the capacity of pre-trained text-layout models in handling multi-page and multi-level challenges.

ECON: On the Detection and Resolution of Evidence Conflicts (2024.emnlp-main)

Copied to clipboard

Cheng Jiayang, Chunkit Chan, Qianqian Zhuang, Lin Qiu, Tianhang Zhang, Tengxiao Liu, Yangqiu Song, Yue Zhang, Pengfei Liu, Zheng Zhang

ProtoCycle: Reflective Tool-Augmented Planning for Text-Guided Protein Design (2026.findings-acl)

Challenge:	Recent studies have shown that AI generated content is more likely to dominate search results, making it difficult to detect when compared to human-produced content.
Approach:	They propose a method for generating diverse, validated evidence conflicts to simulate real-world misinformation scenarios.
Outcome:	The proposed method enables the detection of conflicting information in real-world scenarios and shows that weaker models struggle with similar answer conflicts while stronger models show robust performance.

Copied to clipboard

Yutang Ge, Guojiang Zhao, Sihang Li, Zheng Cheng, Zifeng Zhao, Hanchen Xia, Guolin Ke, Linfeng Zhang, Zhifeng Gao, Yu Guang Wang

Challenge:	Recent deep generative models have already shown encouraging * Equal contribution.
Approach:	They propose to use generic instruction-tuned LLMs as direct text-to-sequence generators to achieve this goal.
Outcome:	Recent studies show that reflection improves sequence quality and alignment while maintaining competitive foldability.

Beyond Itinerary Planning—A Real-World Benchmark for Multi-Turn and Tool-Using Travel Tasks (2026.acl-long)

Copied to clipboard

Xiang Cheng, Yulan Hu, Xiangwen Zhang, Lu Xu, Lide Tan, Zheng Pan, Xin Li, Yong Liu

LAGCL4Rec: When LLMs Activate Interactions Potential in Graph Contrastive Learning for Recommendation (2025.findings-emnlp)

Challenge:	Existing studies on LLM performance on travel planning have shown that existing settings are limited due to limited domain coverage, insufficient modeling of users’ implicit preferences in multi-turn conversations, and a lack of evaluation of agents’ capability boundaries.
Approach:	They propose a benchmark to evaluate LLMs' planning and tool-use abilities in real-world settings by collecting user queries, user preferences, and tools from real scenarios.
Outcome:	The proposed benchmark evaluates agents' capabilities in real-world settings and shows that even advanced models exhibit imbalanced performance across different capabilities.

Copied to clipboard

Leqi Zheng, Chaokun Wang, Canzhi Chen, Jiajun Zhang, Cheng Wu, Zixin Song, Shannan Yan, Ziyang Liu, Hongwei Li

Exploring Speaker-Related Information in Spoken Language Understanding for Better Speaker Diarization (2023.findings-acl)

Challenge:	Traditional contrastive learning methods treat negative feedback as equally hard or easy, ignoring informative semantic difficulty during training.
Approach:	They propose a framework leveraging Large Language Models to Activate interactions in Graph Contrastive Learning for Recommendation.
Outcome:	The proposed framework outperforms state-of-the-art benchmarks on multiple benchmarks.

Copied to clipboard

Luyao Cheng, Siqi Zheng, Zhang Qinglin, Hui Wang, Yafeng Chen, Qian Chen

OneRec-Think: In-Text Reasoning for Generative Recommendation (2026.acl-long)

Challenge:	Current speaker diarization systems consider only acoustic information, resulting in performance degradation when encountering adverse acustic environment.
Approach:	They propose methods to extract speaker-related information from conversational semantics in multi-party meetings.
Outcome:	The proposed method improves on AISHELL-4 and AliMeeting datasets on speakers diarization and speaker-turn detection.

Copied to clipboard

Zhanyu Liu, Shiyao Wang, Xingmei Wang, Rongzhou Zhang, Jiaxin Deng, Honghui Bao, Jinghao Zhang, Wuchao Li, PengFei Zheng, Xiangyu Wu, Yifei Hu, Qigen Hu, Xinchen Luo, Lejian Ren, Zhang Zixing, Qianqian Wang, Kuo Cai, Yunfan Wu, Hongtao Cheng, Zexuan Cheng, Lu Ren, Huanjie Wang, Yi Su, Ruiming Tang, Kun Gai, Guorui Zhou

AraMUS: Pushing the Limits of Data and Model Scale for Arabic Natural Language Processing (2023.findings-acl)

Challenge:	Existing generative models lack the capacity for explicit and controllable reasoning, a key advantage of LLMs.
Approach:	They propose a framework that integrates dialogue, reasoning, and personalized recommendation.
Outcome:	Experiments across public benchmarks show state-of-the-art performance.

Copied to clipboard

Asaad Alghamdi, Xinyu Duan, Wei Jiang, Zhenhai Wang, Yimeng Wu, Qingrong Xia, Zhefeng Wang, Yi Zheng, Mehdi Rezagholizadeh, Baoxing Huai, Peilun Cheng, Abbas Ghaddar

NLP Systems That Can’t Tell Use from Mention Censor Counterspeech, but Teaching the Distinction Helps (2024.naacl-long)

Challenge:	Developing monolingual large Pre-trained Language Models (PLMs) is shown to be very successful in handling different tasks in Natural Language Processing (NLP).
Approach:	They present AraMUS, the largest Arabic PLM with 11B parameters trained on 529GB of high-quality Arabic textual data.
Outcome:	The proposed model achieves state-of-the-art performance on a diverse set of Arabic classification and generative tasks.

Copied to clipboard

Kristina Gligoric, Myra Cheng, Lucia Zheng, Esin Durmus, Dan Jurafsky