Papers by Cheng Deng

LLMs Assist NLP Researchers: Critique Paper (Meta-)Reviewing (2024.emnlp-main)

Challenge:	Existing LLM-based recommender systems rely on standard fine-tuning methodologies, often ignoring hallucination issues during the fine-uning process.
Approach:	They propose a logit space constraint-based fine-tuning framework to mitigate hallucination in LLM-based recommenders by incorporating Kullback–Leibler divergence into the training objective.
Outcome:	Experiments on two recommendation models with distinct LLM backbones and four real-world datasets show that LCFT reduces hallucination and enhances recommendation performance.

Copied to clipboard

Jiangshu Du, Yibo Wang, Wenting Zhao, Zhongfen Deng, Shuaiqi Liu, Renze Lou, Henry Zou, Pranav Narayanan Venkit, Nan Zhang, Mukund Srinath, Haoran Zhang, Vipul Gupta, Yinghui Li, Tao Li, Fei Wang, Qin Liu, Tianlin Liu, Pengzhi Gao, Congying Xia, Chen Xing, Cheng Jiayang, Zhaowei Wang, Ying Su, Raj Shah, Ruohao Guo, Jing Gu, Haoran Li, Kangda Wei, Zihao Wang, Lu Cheng, Surangika Ranathunga, Meng Fang, Jie Fu, Fei Liu, Ruihong Huang, Eduardo Blanco, Yixin Cao, Rui Zhang, Philip Yu, Wenpeng Yin

Challenge:	a comparative analysis of paper (meta-)reviews by large language models (LLMs) aims to identify and distinguish LLMs from human activities .
Approach:	They present a comparative analysis to identify and distinguish LLM activities from human activities.
Outcome:	The proposed analysis aims to improve recognition of instances when someone implicitly uses LLMs for reviewing activities.

G2RPO-A: Guided Group Relative Policy Optimization with Adaptive Guidance (2026.acl-long)

Copied to clipboard

Yongxin Guo, Wenbo Deng, Zhenglin Cheng, Xiaoying Tang

UHGEval: Benchmarking the Hallucination of Chinese Large Language Models via Unconstrained Generation (2024.acl-long)

Challenge:	Recent advances in reasoning-centric large language models (LLMs) have significantly expanded the performance boundaries of LLMs, showcasing the immense potential of reasoning-enhanced models.
Approach:	They propose an adaptive algorithm that injects ground-truth reasoning steps into roll-out trajectories to compensate for SLMs’ inherent weaknesses.
Outcome:	Experiments on mathematical reasoning and code-generation benchmarks confirm that G2RPO-A substantially outperforms vanilla GRPO.

Copied to clipboard

Xun Liang, Shichao Song, Simin Niu, Zhiyu Li, Feiyu Xiong, Bo Tang, Yezhaohui Wang, Dawei He, Cheng Peng, Zhonghao Wang, Haiying Deng

Revealing and Enhancing Core Visual Regions: Harnessing Internal Attention Dynamics for Hallucination Mitigation in LVLMs (2026.findings-acl)

Challenge:	Large language models (LLMs) produce hallucinated text, compromising their practical utility in professional contexts.
Approach:	They have developed an unconstrained hallucination generation evaluation benchmark that contains hallucines generated by large language models with minimal restrictions.
Outcome:	The proposed benchmarks are based on a Chinese-language dataset that is lacking in the field.

Copied to clipboard

Guangtao Lyu, Qi Liu, Chenghao Xu, Jiexi Yan, Muli Yang, Xueting Li, Fen Fang, Cheng Deng

Tell Me More! Towards Implicit User Intention Understanding of Language Model Driven Agents (2024.acl-long)

Challenge:	Existing training-free methods are vulnerable to the attention sink phenomenon . Existing methods include contrastive decoding and auxiliary expert models .
Approach:	They propose a training-free attention intervention that constructs a PAD map to identify semantically core visual regions and applies per-head Median Absolute Deviation Scaling to adaptively control the intervention strength.
Outcome:	The proposed intervention improves visual grounding and reduces hallucinations on multiple LVLMs and benchmarks.

Copied to clipboard

Cheng Qian, Bingxiang He, Zhong Zhuang, Jia Deng, Yujia Qin, Xin Cong, Zhong Zhang, Jie Zhou, Yankai Lin, Zhiyuan Liu, Maosong Sun

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

Challenge:	Current language model-driven agents lack mechanisms for effective user participation, which is crucial given the vagueness commonly found in user instructions.
Approach:	They propose a benchmark to inspect users’ implicit intentions through explicit queries and a model expert as the upstream in agent design to enhance user-agent interaction.
Outcome:	The proposed approach excels at identifying vague user tasks, recovering and summarizing critical missing information, setting precise and necessary agent execution goals, and minimizing redundant tool usage, thus boosting overall efficiency.

Copied to clipboard

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

MLaKE: Multilingual Knowledge Editing Benchmark for Large Language Models (2025.coling-main)

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

Zihao Wei, Jingcheng Deng, Liang Pang, Hanxing Ding, Huawei Shen, Xueqi Cheng

InstructSafety: A Unified Framework for Building Multidimensional and Explainable Safety Detector through Instruction Tuning (2023.findings-emnlp)

Challenge:	Existing studies on knowledge editing focus on monolingual scenarios, neglecting the complexities presented by multilingual contexts and multi-hop reasoning.
Approach:	They propose a benchmark to evaluate the adaptability of multilingual knowledge editing methods.
Outcome:	The proposed benchmark evaluates the adaptability of multilingual knowledge editing methods across five languages.

Copied to clipboard

Zhexin Zhang, Jiale Cheng, Hao Sun, Jiawen Deng, Minlie Huang

Responsible & Trustworthy NLP

Challenge:	Existing safety detection systems have limitations in terms of their versatility and interpretability.
Approach:	They introduce a safety detection framework that unifies 7 common sub-tasks into a uniform formulation and process 39 human-annotated datasets for instruction tuning.
Outcome:	The proposed framework unifies 7 common sub-tasks into a uniform formulation and then runs on 39 human-annotated datasets to fine-tune it.

DialogGen: Multi-modal Interactive Dialogue System with Multi-turn Text-Image Generation (2025.findings-naacl)

Copied to clipboard

Minbin Huang, Yanxin Long, Xinchi Deng, Ruihang Chu, Jiangfeng Xiong, Xiaodan Liang, Hong Cheng, Qinglin Lu, Wei Liu

Curriculum Consistency Learning for Conditional Sentence Generation (2024.emnlp-main)

Challenge:	Text-to-image (T2I) generation models have advanced in recent years, but effective interaction with these models is challenging for average users due to the need for specialized prompt engineering knowledge and the inability to perform multi-turn image generation.
Approach:	They propose to use off-the-shelf MLLMs and T2I models to build a multi-modal interactive dialogue system (MIDS) that can generate correct output modalities and coherence of output images.
Outcome:	The proposed pipeline can generate correct output modalities and coherent multi-modal outputs compared with other state-of-the-art models.

Copied to clipboard

Liangxin Liu, Xuebo Liu, Lian Lian, Shengjun Cheng, Jun Rao, Tengfei Yu, Hexuan Deng, Min Zhang

Text Generation

Challenge:	Consistency learning (CL) has proven to be a valuable technique for improving the robustness of conditional sentence generation models.
Approach:	They propose a strategy that guides models to learn consistency in alignment with their current capacity to differentiate between features.
Outcome:	The proposed strategy delivers +2.0 accuracy point improvement compared with vanilla IT and +0.7 COMET scores over traditional CL methods in MT tasks.

Constructing Highly Inductive Contexts for Dialogue Safety through Controllable Reverse Generation (2022.findings-emnlp)

Copied to clipboard

Zhexin Zhang, Jiale Cheng, Hao Sun, Jiawen Deng, Fei Mi, Yasheng Wang, Lifeng Shang, Minlie Huang

Dialogue Systems & Conversational Agents

Challenge:	Existing methods to detect toxic generation of pretrained language models rely on templates, data extraction, crowdsourcing workers or automatic generation.
Approach:	They propose a method to construct adversarial contexts conditioned on a given response . they augment existing dataset BAD+ and construct a new dataset B AD+ .
Outcome:	The proposed method can detect toxic or biased content in large pretrained language models.

VisPCO: Visual Token Pruning Configuration Optimization via Budget-Aware Pareto-Frontier Learning for Vision-Language Models (2026.acl-long)

Copied to clipboard

Huawei Ji, Yuanhao Sun, Yuan Jin, Cheng Deng, Jiaxin Ding, Luoyi Fu, Xinbing Wang

Visual Data in NLP

Challenge:	Existing methods for visual token pruning rely on predefined configurations without determining whether they achieve optimal performance.
Approach:	They propose a framework that formulates visual token pruning as a Pareto configuration optimization problem to automatically identify optimal configurations.
Outcome:	The proposed framework approximates the empirical Pareto frontier obtained through grid search and generalizes well across pruning methods and VLM architectures.

XY-Tokenizer: Mitigating the Semantic-Acoustic Conflict in Low-Bitrate Speech Codecs (2026.acl-long)

Copied to clipboard

Yitian Gong, Luozhijie Jin, Kuangwei Chen, Dong Zhang, Ruifan Deng, Xiaogui Yang, Xin Zhang, Zhaoye Fei, Qinyuan Cheng, Shimin Li, Xipeng Qiu

Logic-Thinker: Teaching Large Language Models to Think more Logically. (2025.findings-emnlp)

Challenge:	Existing speech codecs struggle to balance these objectives at low bitrates . XY-Tokenizer achieves stronger semantic alignment than representative semantic-distillation codec .
Approach:	They propose a low-bitrate speech codec that aligns discrete speech representations with text while preserving fine-grained acoustic details for reconstruction.
Outcome:	The proposed codec outperforms existing low-bitrate speech codecs in speech understanding and generation tasks.

Copied to clipboard

Chengyao Wen, Qiang Cheng, Shaofei Wang, Zhizhen Liu, Deng Zhao, Lei Liang

NegotiationToM: A Benchmark for Stress-testing Machine Theory of Mind on Negotiation Surrounding (2024.findings-emnlp)

Challenge:	Recent Large Reasoning Models (LRMs) have demonstrated the ability to generate long chains of thought (LongCoT) LongCoT still faces challenges such as redundancy and logical incoherence.
Approach:	They propose a neural-symbolic reasoning framework that generates chains of thought . they propose Logic-Thinker, which transforms symbolic solvers into chains of thoughts .
Outcome:	The proposed framework outperforms models fine-tuned with ThinkerCoT on logic reasoning tasks.

Copied to clipboard

Chunkit Chan, Cheng Jiayang, Yauwai Yim, Zheye Deng, Wei Fan, Haoran Li, Xin Liu, Hongming Zhang, Weiqi Wang, Yangqiu Song

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

Challenge:	Theory of mind evaluations currently focus on testing models using machine-generated data or game settings prone to shortcuts and spurious correlations.
Approach:	They propose a benchmark to stress-test machine ToM in real-world negotiation surrounding covered multi-dimensional mental states.
Outcome:	The proposed benchmark builds upon the Belief-Desire-Intention theory and conducts the necessary empirical experiments to evaluate large language models.

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.

Editing Large Language Models: Problems, Methods, and Opportunities (2023.emnlp-main)

Copied to clipboard

Yunzhi Yao, Peng Wang, Bozhong Tian, Siyuan Cheng, Zhoubo Li, Shumin Deng, Huajun Chen, Ningyu Zhang

GR1: Reinforcement-Enhanced LLM for Geoscience Reasoning (2026.findings-acl)

Challenge:	Recent advances in model editing for LLMs have created challenges and opportunities for the community.
Approach:	They propose to alter the behavior of LLMs efficiently within a specific domain without negatively impacting performance across other inputs.
Outcome:	The proposed method alters behavior of LLMs efficiently within a specific domain without negatively impacting performance across other inputs.

Copied to clipboard

Yule Xie, Jiaxin Ding, Cheng Deng, Shiqing Gao, Junran Zhang, Sibo Zhang, Zeyuan Wang, Ke Wu, Xin Ding, Luoyi Fu, Meng Jin, Xinbing Wang

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

Challenge:	Recent advances in large language models have demonstrated RL's substantial capacity to enhance multi-step reasoning beyond what supervised instruction tuning achieves.
Approach:	They propose a framework that converts multimodal questions into descriptive text . they propose RL-enhanced geoscience reasoning that can be fine-tuned to a text-only level .
Outcome:	The proposed framework improves accuracy and accuracy on multimodal questions while preserving answerability and difficulty.

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.

Exploiting Intrinsic Multilateral Logical Rules for Weakly Supervised Natural Language Video Localization (2024.acl-long)

Copied to clipboard

Zhe Xu, Kun Wei, Xu Yang, Cheng Deng

XToM: Exploring the Multilingual Theory of Mind for Large Language Models (2026.acl-long)

Challenge:	Existing methods for WS-NLVL rarely consider complex temporal relations enclosing the language query, yielding illogical predictions.
Approach:	They propose a plug-and-play method to exploit temporal relations and logical rules for WS-NLVL.
Outcome:	The proposed method is able to retrieve the moment corresponding to a language query in a video with only video-language pairs utilized during training.

Copied to clipboard

Chunkit Chan, Yauwai Yim, Hongchuan Zeng, Zhiying Zou, Xinyuan Cheng, Zhifan Sun, Zheye Deng, Kawai Chung, Yuzhuo Ao, Fan Yixiang, Cheng Jiayang, Ercong Nie, Ginny Wong, Helmut Schmid, Hinrich Schuetze, Simon See, Yangqiu Song

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

Challenge:	Existing evaluations of ToM in LLMs are limited to English, neglecting the linguistic diversity that shapes human cognition.
Approach:	They propose a multilingual benchmark that evaluates ToM across five languages . they find that models excel in multilingual language understanding, but their ToM performance varies across languages.
Outcome:	The proposed benchmark evaluates LLMs across five languages and incorporates diverse task scenarios.

Copied to clipboard

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

The Right Time Matters: Data Arrangement Affects Zero-Shot Generalization in Instruction Tuning (2025.findings-acl)

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

Bingxiang He, Ning Ding, Cheng Qian, Jia Deng, Ganqu Cui, Lifan Yuan, Haiwen Hong, Huan-ang Gao, Longtao Huang, Hui Xue, Huimin Chen, Zhiyuan Liu, Maosong Sun

Semantically Comprehensive Token Pruning in LVLMs via Maximizing Concept Coverage (2026.acl-long)

Challenge:	Existing work on instruction tuning has focused on task level, without considering that tasks are artificially defined and, to LLMs, merely consist of tokens and representations.
Approach:	They propose a training data arrangement framework that allows for continual learning and loss reduction.
Outcome:	The proposed framework promotes continual learning and loss reduction on unseen tasks.

Copied to clipboard

Xueting Li, Qi Liu, Chenghao Xu, Xu Yang, Guangtao Lyu, Jiahua Li, Cheng Deng

Explainability & Interpretability in NLP

Challenge:	Existing visual token pruning methods leverage simple metrics derived from human experience, such as attention or similarity, to rank and select tokens within a highly entangled feature space.
Approach:	They propose a novel visual token pruning method that uses a concept-driven paradigm to quantify the Marginal Semantic Gain of each token's contribution to uncovered concepts.
Outcome:	The proposed method outperforms state-of-the-art methods in a concept-driven model while maintaining semantic completeness.

When Audio and Text Disagree: Revealing Text Bias in Large Audio-Language Models (2025.emnlp-main)

Copied to clipboard

Cheng Wang, Gelei Deng, Xianglin Yang, Han Qiu, Tianwei Zhang

Distance between Relevant Information Pieces Causes Bias in Long-Context LLMs (2025.findings-acl)

Challenge:	Large Audio-Language Models (LALMs) are augmented with the ability to perceive audio, but their reliability when faced with conflicting inputs remains largely unexplored.
Approach:	They examine how LALMs prioritize information when presented with inconsistent audio-text pairs.
Outcome:	The proposed models display a significant bias toward textual input when presented with inconsistent audio-text pairs.

Copied to clipboard

Runchu Tian, Yanghao Li, Yuepeng Fu, Siyang Deng, Qinyu Luo, Cheng Qian, Shuo Wang, Xin Cong, Zhong Zhang, Yesai Wu, Yankai Lin, Huadong Wang, Xiaojiang Liu

Challenge:	Positional biases in large language models hinder their ability to process long inputs.
Approach:	They propose a benchmark to assess positional bias in large language models involving multiple pieces of relevant information.
Outcome:	The proposed benchmark assesses the performance of long-context language models by examining their models with different input lengths and tasks.

RecMem: Recurrence-based Memory Consolidation for Efficient and Effective Long-Running LLM Agents (2026.findings-acl)

Copied to clipboard

Zijie Dai, Shiyuan Deng, Sheng Guan, Yizhou Tian, Xin Yao, Xiao Yan, James Cheng

Green & Sustainable NLP

Challenge:	Existing memory systems invoke LLMs to extract episodic and semantic memory, and this leads to substantial token consumption.
Approach:	They propose a method that stores incoming interactions in a subconscious memory layer and encodes them using lightweight embedding models for retrieval.
Outcome:	Experiments show that RecMem reduces the memory construction token cost of three SOTA memory systems by up to 87% while exceeding their accuracy.

Projecting Out the Malice: A Global Subspace Approach to LLM Detoxification (2026.acl-long)

Copied to clipboard

Zenghao Duan, Zhiyi Yin, Zhichao Shi, Liang Pang, Shaoling Jing, Zihe Huang, Jiayi Wu, Yu Yan, Jingcheng Deng, Huawei Shen, Xueqi Cheng

Challenge:	Large language models (LLMs) exhibit exceptional performance but pose inherent risks of generating toxic content.
Approach:	They propose a method that removes toxic subspaces from FFN parameters . they propose to use a lightweight method to eliminate toxic subespaces .
Outcome:	The proposed method achieves SOTA detoxification while preserving general capabilities without large-scale retraining.

Beyond Binary Preferences: Semi-Online Label-Free GRACE-KTO with Group-Wise Adaptive Calibration for High-Quality Long-Text Generation (2025.findings-emnlp)

Copied to clipboard

Jingyang Deng, Ran Chen, Jo-Ku Cheng, Jinwen Ma

Text Generation

Challenge:	Generating high-quality long-text remains challenging for Large Language Models (LLMs), as conventional supervised fine-tuning fails to ensure overall quality due to its teacher-forcing nature.
Approach:	They propose a semi-online framework that transforms KTO’s binary signals into dynamically calibrated intra-group rewards.
Outcome:	The proposed framework transforms binary signals into dynamically calibrated intra-group rewards.

Rotation Control Unlearning: Quantifying and Controlling Continuous Unlearning for LLM with The Cognitive Rotation Space (2026.findings-acl)

Copied to clipboard

Xiang Zhang, Kun Wei, Xu Yang, Jiahua Li, Su Yan, Cheng Deng

Challenge:	Existing methods to remove undesirable data from Large Language Models suffer from cumulative catastrophic utility loss under continuous unlearning requests.
Approach:	They propose a method that leverages the rotational salience weight of RCU to quantify and control the unlearning degree in the continuous unlearning process.
Outcome:	The proposed method achieves SOTA performance without a retained dataset.

LLM Knows Body Language, Too: Translating Speech Voices into Human Gestures (2024.acl-long)

Copied to clipboard

Chenghao Xu, Guangtao Lyu, Jiexi Yan, Muli Yang, Cheng Deng

Hierarchical Reinforcement Learning with Augmented Step-Level Transitions for LLM Agents (2026.acl-long)

Challenge:	despite advances in the generation of realistic human gestures, the process often includes unintended, meaningless, or non-realistic gestures.
Approach:	They propose a framework that leverages large language models to generate human gestures . the primary stage employs a transformer-based auto-encoder network to encode human gesture into discrete symbols .
Outcome:	The proposed framework has demonstrated state-of-the-art performance on public TED and TED-Expressive datasets.

Copied to clipboard

Shuai Zhen, Yanhua Yu, Ruopei Guo, Nan Cheng, Yang Deng

Challenge:	Existing Large language model agents rely on increasingly long interaction histories, resulting in high computational cost and limited scalability.
Approach:	They propose a hierarchical reinforcement learning framework that enables step-level learning by conditioning only on single-step transitions rather than full interaction histories.
Outcome:	The proposed framework outperforms baselines on ScienceWorld and ALFWorld benchmarks in terms of performance and generalization while reducing token usage.

ODL-TempLLM: Ontology-Guided and Description Logic-Reasoned Temporal Reasoning with LLMs (2026.acl-long)

Copied to clipboard

Jinshuo Liu, Cheng Bi, Meng Wang, Juan Deng, Donghong Ji, Jeff Z. Pan

SeaLLMs - Large Language Models for Southeast Asia (2024.acl-demos)

Challenge:	Temporal reasoning is crucial for large language models to understand event concurrency and complex temporal interactions in natural language.
Approach:	They propose an ontology-guided and description logic–constrained temporal reasoning paradigm that shifts focus from internal inference to the explicit modeling of temporal structure.
Outcome:	The proposed method outperforms state-of-the-art methods by 2.07–31.83 F1 points and 1.00–30.73 EM points, exhibiting strong generalization and highlighting the potential of explicit temporal reasoning.

Copied to clipboard

Xuan-Phi Nguyen, Wenxuan Zhang, Xin Li, Mahani Aljunied, Zhiqiang Hu, Chenhui Shen, Yew Ken Chia, Xingxuan Li, Jianyu Wang, Qingyu Tan, Liying Cheng, Guanzheng Chen, Yue Deng, Sen Yang, Chaoqun Liu, Hang Zhang, Lidong Bing

The Evolution of Thought: Tracking LLM Overthinking via Reasoning Dynamics Analysis (2026.acl-long)

Challenge:	Existing large language models favor high-resource languages, such as English, at the expense of low-resourced and regional languages.
Approach:	They propose a series of language models that specifically focuses on Southeast Asian languages.
Outcome:	SeaLLM models outperform ChatGPT-3.5 in non-Latin languages by large margins . linguistic disparity impedes access to state-of-the-art AI technologies for non-English-speaking populations .

Copied to clipboard

Zihao Wei, Liang Pang, Jiahao Liu, Wenjie Shi, Jingcheng Deng, Shicheng Xu, Zenghao Duan, Jingang Wang, Fei Sun, Huawei Shen, Xueqi Cheng

NewsBench: A Systematic Evaluation Framework for Assessing Editorial Capabilities of Large Language Models in Chinese Journalism (2024.acl-long)

Challenge:	Explicit reasoning trajectories increase performance but often trigger overthinking . despite its importance, this study examines how each step of reasoning affects the final outcome .
Approach:	They propose a Reasoning Completion Point Detector that detects the RCP by monitoring rank dynamics of termination tokens.
Outcome:	The proposed method reduces token usage by up to 44% while preserving accuracy.

Copied to clipboard

Miao Li, Ming-Bin Chen, Bo Tang, ShengbinHou ShengbinHou, Pengyu Wang, Haiying Deng, Zhiyu Li, Feiyu Xiong, Keming Mao, Cheng Peng, Yi Luo

Following the Autoregressive Nature of LLM Embeddings via Compression and Alignment (2025.emnlp-main)

Challenge:	a novel evaluation framework assesses the capabilities of Large Language Models (LLMs) for editorial capabilities in Chinese journalism.
Approach:	They propose to use a benchmark dataset to assess the capabilities of Large Language Models (LLMs) for editorial capabilities in Chinese journalism.
Outcome:	The proposed evaluation framework is based on a dataset of 1,267 test samples in 24 news domains.

Copied to clipboard

Jingcheng Deng, Zhongtao Jiang, Liang Pang, Zihao Wei, Liwei Chen, Kun Xu, Yang Song, Huawei Shen, Xueqi Cheng

UPRISE: Universal Prompt Retrieval for Improving Zero-Shot Evaluation (2023.emnlp-main)

Challenge:	Experimental results demonstrate that our method significantly outperforms traditional contrastive learning approaches when using the same amount of data.
Approach:	They propose a new contrastive learning method built on embedding conditional probability distributions that integrates two tasks: information compression and conditional distribution alignment.
Outcome:	The proposed method outperforms traditional contrastive learning approaches and achieves comparable performance to state-of-the-art models when using the same amount of data.

Copied to clipboard

Daixuan Cheng, Shaohan Huang, Junyu Bi, Yuefeng Zhan, Jianfeng Liu, Yujing Wang, Hao Sun, Furu Wei, Weiwei Deng, Qi Zhang

Dual Activation-Weight Sparsity: A Training-Free Framework for Efficient Large Language Model Compression (2026.acl-long)

Challenge:	Large Language Models (LLMs) have impressive capabilities but need for task-specific prompt engineering can hinder their generalization.
Approach:	They propose a lightweight and versatile retriever that automatically retrieves prompts for a given zero-shot task input.
Outcome:	The proposed model is universally applicable across tasks and models . it mitigates hallucination problem in chatGPT, and it improves even the strongest LLMs.

Copied to clipboard

Luoyang Sun, Guangyan Li, Cheng Deng, Haifeng Zhang, Jian Zhao, Yongqiang Tang, Wensheng Zhang, Jun Wang

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

Challenge:	Large language models (LLMs) excel at natural language tasks but face deployment bottlenecks due to computational demands.
Approach:	They propose a training-free framework that exploits activation and weight sparsity . they use a three-tier routing strategy that uses magnitude-based pruning .
Outcome:	Experiments on Llama and Mistral models show that DAWS outperforms activation-weight sparsity pruning methods.

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

Specialist or Generalist? Instruction Tuning for Specific NLP Tasks (2023.emnlp-main)

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

Chufan Shi, Yixuan Su, Cheng Yang, Yujiu Yang, Deng Cai

Logical Structure as Knowledge: Enhancing LLM Reasoning via Structured Logical Knowledge Density Estimation (2026.findings-acl)

Challenge:	Recent studies have shown that instruction tuning can be a data-efficient method for transforming large language models into generalist models, but their performance lags behind specialist models trained exclusively for specific tasks.
Approach:	They propose to incorporate broadcoverage generalist instruction tuning into large language models to build a specialist model by incorporating task specificity and skill requirements.
Outcome:	The proposed method improves model performance when task coverage is broad and when training data is limited.

Copied to clipboard

Zhen Bi, Zhenlin Hu, Xueshu Chen, Mingyang Chen, Cheng Deng, Yida Xue, Zhen Wang, Qing Shen, Ningyu Zhang, Jungang Lou

RegaVAE: A Retrieval-Augmented Gaussian Mixture Variational Auto-Encoder for Language Modeling (2023.findings-emnlp)

Challenge:	Existing data-centric paradigms equate quality with factuality or diversity and ignore the internal logical complexity of training samples.
Approach:	They propose a density-aware re-cognizing optimization strategy that prioritizes high-density logical samples to align training with the model's reasoning boundary.
Outcome:	The proposed metric outperforms existing methods and improves reasoning performance without increasing total data volume.

Copied to clipboard

Jingcheng Deng, Liang Pang, Huawei Shen, Xueqi Cheng

Fisher-Driven Adaptive Locating for Knowledge Editing in Large Language Models (2026.acl-long)

Challenge:	Existing research on retrieval-augmented language models has two main problems: determining what information to retrieve and effectively combining retrieved information during generation.
Approach:	They propose a retrieval-augmented language model that captures current and future information from source and target text into a latent space.
Outcome:	The proposed model is more efficient than explicit raw text, but limited by context length and noise.

Copied to clipboard

Chenghao Xu, Jiexi Yan, Guangtao Lyu, Qi Liu, Muli Yang, Cheng Deng