Papers by Jie He

Semantics of the Unwritten: The Effect of End of Paragraph and Sequence Tokens on Text Generation with GPT2 (2021.acl-srw)

Challenge:	Existing methods for information extraction (IE) focus on training task-specific models, while common knowledge among different IE tasks is not explicitly modeled.
Approach:	They propose a regularization-based transfer learning method for IE via an instructed graph decoder which decodes various complex structures into a graph uniformly based on corresponding instructions.
Outcome:	The proposed method can learn common knowledge from existing datasets and transfer it to a new dataset with new labels.

Copied to clipboard

He Bai, Peng Shi, Jimmy Lin, Luchen Tan, Kun Xiong, Wen Gao, Jie Liu, Ming Li

Boosting Large Language Models with Continual Learning for Aspect-based Sentiment Analysis (2024.findings-emnlp)

Challenge:	Experimental results show that pre-trained language model GPT2 can generate better continuations by learning to generate the in the fine-tuning stage.
Approach:	They conduct experiments on an English essay dataset using Chinese-GPT2 . they find that the model can generate better continuations by learning to generate the in the fine-tuning stage.
Outcome:	The pre-trained language model GPT2 can generate better continuations by learning to generate the in the fine-tuning stage.

Copied to clipboard

Xuanwen Ding, Jie Zhou, Liang Dou, Qin Chen, Yuanbin Wu, Arlene Chen, Liang He

Sentiment Analysis

Challenge:	Existing studies focus on improving the performance of domain-specific models based on the target dataset.
Approach:	They propose a Large Language Model-based Continual Learning (LLM-CL) model for ABSA that learns the target domain’s ability while maintaining the history domains’ abilities.
Outcome:	The proposed model obtains new state-of-the-art over 19 datasets.

FlowRAG: Synergizing Explicit Reasoning via Frequency-Aware Multi-Granularity Graph Flow (2026.findings-acl)

Copied to clipboard

Bihao Zhan, Zongsheng Cao, Jie Zhou, Bo Zhang, Liang He

Information Retrieval

Challenge:	Existing methods for Graph-based retrieval-augmented generation rely on implicit semantic relevance propagation.
Approach:	They propose a semantic-aware retrieval framework that improves both semantic recall and explicit reasoning.
Outcome:	Extensive experiments show that FlowRAG improves both semantic recall and explicit reasoning.

Contrastive Preference Learning for Neural Machine Translation (2024.findings-naacl)

Copied to clipboard

Jianfei He, Shichao Sun, Sen Peng, Jie Xu, Xiaohua Jia, Wenjie Li

Machine Translation

Challenge:	Existing discrepancies between token-level objective and overall sequence-level quality of a model are causing exposure bias and other issues in NMT.
Approach:	They propose a contrastive preference model that integrates an indicator function to fine-tune a pre-trained model in Neural Machine Translation.
Outcome:	The proposed model outperforms the traditional Plackett-Luce model on three language pairs and also outperFORMs token-level and sequence-level baseline models.

Optimizing Question Semantic Space for Dynamic Retrieval-Augmented Multi-hop Question Answering (2025.acl-long)

Copied to clipboard

Linhao Ye, Lang Yu, Zhikai Lei, Qin Chen, Jie Zhou, Liang He

Information Retrieval

Challenge:	Existing retrieval-augmented generation methods are insufficient for multi-hop question answering . however, they tend to generate hallucinations due to semantic mismatching .
Approach:	They propose to optimize question semantic space for dynamic retrieval-augmented multi-hop question answering by optimizing the semantic embeddings.
Outcome:	The proposed method outperforms existing RAG methods in both in- and out-of-domain settings.

Cross-layer Attention Sharing for Pre-trained Large Language Models (2026.tacl-1)

Copied to clipboard

Yongyu Mu, Yuzhang Wu, Yuchun Fan, Chenglong Wang, Hengyu Li, Jiali Zeng, Qiaozhi He, Murun Yang, Fandong Meng, Jie Zhou, Tong Xiao, Jingbo Zhu

Unregulated Chinese-to-English Data Expansion Does NOT Work for Neural Event Detection (2022.coling-1)

Challenge:	Existing studies focus on compressing the Key-Value cache or grouping attention heads, while overlooking redundancy between layers.
Approach:	They propose a lightweight substitute for self-attention in well-trained LLMs that uses feed-forward networks to align attention heads between adjacent layers and low-rank matrices to approximate differences in layer-wise attention weights.
Outcome:	The proposed model reduces redundancy by sharing weights across layers while maintaining high response quality while reducing redundant calculations within 53% 84% of the total layers.

Copied to clipboard

Zhongqiu Li, Yu Hong, Jie Wang, Shiming He, Jianmin Yao, Guodong Zhou

Multilinguality

Challenge:	Experimental results show that cross-language data expansion results in performance degradation.
Approach:	They leverage cross-language data expansion and retraining to enhance neural Event Detection on English ACE corpus.
Outcome:	The proposed method improves ED performance by 1.6% over the straight data combination.

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

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

UniArk: Improving Generalisation and Consistency for Factual Knowledge Extraction through Debiasing (2024.naacl-long)

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

Yijun Yang, Jie He, Pinzhen Chen, Victor Gutierrez Basulto, Jeff Pan

ChunQiuTR: Time-Keyed Temporal Retrieval in Classical Chinese Annals (2026.findings-acl)

Challenge:	Existing studies have investigated the potential of language models as knowledge bases and the existence of severe biases when extracting factual knowledge.
Approach:	They propose an adapter-based framework for generalised factual knowledge extraction using simple methods without introducing extra parameters.
Outcome:	The proposed framework improves the model’s out-of-domain generalisation and consistency under various prompts.

Copied to clipboard

Yihao Wang, Zijian He, Jie Ren, Keze Wang

A Multi-Format Transfer Learning Model for Event Argument Extraction via Variational Information Bottleneck (2022.coling-1)

Challenge:	Historical research often focuses on finding exact record for a specific regnal month . classical Chinese sources are a canonical example of evidence-centric retrieval .
Approach:	They propose a time-keyed retrieval benchmark that organizes records by month-level reign keys . they propose 'CTD', a dual-encoder that combines absolute context with offset biasing .
Outcome:	The proposed benchmark organizes records by month-level reign keys and includes chrono-near confounders that mimic real retrieval failures.

Copied to clipboard

Jie Zhou, Qi Zhang, Qin Chen, Qi Zhang, Liang He, Xuanjing Huang

Contrastive Zero-Shot Learning for Cross-Domain Slot Filling with Adversarial Attack (2020.coling-main)

Challenge:	Event argument extraction (EAE) aims to extract arguments with given roles from texts.
Approach:	They propose a multi-format transfer learning model with variational information bottleneck to learn from existing datasets.
Outcome:	The proposed model improves on three benchmark datasets and obtains state-of-the-art performance on EAE.

Copied to clipboard

Keqing He, Jinchao Zhang, Yuanmeng Yan, Weiran Xu, Cheng Niu, Jie Zhou

HyperMoE: Towards Better Mixture of Experts via Transferring Among Experts (2024.acl-long)

Challenge:	Existing approaches to zero-shot slot filling ignore constraints in the latent space and lack robustness.
Approach:	They propose a Contrastive Zero-Shot Learning with Adversarial Attack method for slot filling . they propose to map slot value contextual representations to slot description representations .
Outcome:	The proposed method outperforms state-of-the-art models under zero-shot and few-shot settings.

Copied to clipboard

Hao Zhao, Zihan Qiu, Huijia Wu, Zili Wang, Zhaofeng He, Jie Fu

DiaHalu: A Dialogue-level Hallucination Evaluation Benchmark for Large Language Models (2024.findings-emnlp)

Challenge:	Existing methods for enhancing performance through increased use of expert knowledge often result in diminishing sparsity during expert selection.
Approach:	They propose a framework that integrates the computational processes of MoE with the concept of knowledge transferring in multi-task learning.
Outcome:	The proposed framework outperforms existing methods under identical conditions concerning the number of experts.

Copied to clipboard

Kedi Chen, Qin Chen, Jie Zhou, He Yishen, Liang He

Dialogue Systems & Conversational Agents

Challenge:	Existing benchmarks for hallucination detection are intentionally generated by large language models (LLMs) however, many focus on factuality while ignoring faithfulness.
Approach:	They propose a dialogue-level hallucination evaluation benchmark for large language models . they integrate the topic into prompts and facilitate a dialog between two LLMs .
Outcome:	The proposed benchmark covers four common multi-turn dialogue domains and five hallucination subtypes, extended from factuality and faithfulness hallucines.

Let’s Rectify Step by Step: Improving Aspect-based Sentiment Analysis with Diffusion Models (2024.lrec-main)

Copied to clipboard

Shunyu Liu, Jie Zhou, Qunxi Zhu, Qin Chen, Qingchun Bai, Jun Xiao, Liang He

Sentiment Analysis

Challenge:	Empirical evaluations conducted on eight benchmark datasets underscore the compelling advantages offered by DiffusionABSA when compared against robust baseline models.
Approach:	They propose a diffusion model which extracts aspects step by step and learns a denoising process that progressively restores them in a reverse manner.
Outcome:	Empirical evaluations on eight benchmark datasets underscore the compelling advantages offered by DiffusionABSA when compared against robust baseline models.

Cognitive Scaffold: From Fluid Context to Crystallized Memory for Long-Horizon DeepResearch Agents (2026.acl-long)

Copied to clipboard

Qiuyuan Ai, Zenghuang Fu, Zhaoyang Li, Ping Jiang, Haoyu Wu, Jie Song, Guannan He

Reasoning

Challenge:	Scaling LLM-based agents to long-horizon deep research is constrained by context-noise trade-off . solving a single query may require hundreds of interactions with noisy environments .
Approach:	They propose a factorized memory architecture that decouples the cognitive state into a Fluid Working Context for immediate reasoning and a persistent Knowledge Graph for long-term retention.
Outcome:	The Cognitive Scaffold outperforms baselines on Xbench-DeepSearch, BrowseComp-ZH, and GAIA . it achieves 74.7% Avg@3 and 87.0% Pass@3 on xbench, browseComp, and 88.3% Pass@3.

BUCA: A Binary Classification Approach to Unsupervised Commonsense Question Answering (2023.acl-short)

Copied to clipboard

Jie He, Simon U, Victor Gutierrez-Basulto, Jeff Pan

Information Retrieval

Challenge:	Existing methods for commonsense reasoning use knowledge graphs to train models . however, it is not always possible to have relevant training data available .
Approach:	They propose to transform a question-answer task into a binary classification task by ranking all candidate answers according to their reasonableness.
Outcome:	The proposed approach is less data hungry than existing methods using KGs.

VENUS: A VLLM-driven Video Content Discovery System for Real Application Scenarios (2025.emnlp-industry)

Copied to clipboard

Minyi Zhao, Yi Liu, Jianfeng Wen, Boshen Zhang, Hailang Chang, Zhiheng Ouyang, Jie Wang, Wensong He, Shuigeng Zhou

Visual Data in NLP

Challenge:	Video Content Discovery (VCD) is to identify specific videos defined by a pre-specified text policy.
Approach:	They propose a Vision-Language Large Model-driven video content discovery system called VENUS to solve these problems.
Outcome:	The proposed system generates high-quality, VCD-specific data for model training and extends it to support it better.

LongCLI-Bench: A Preliminary Benchmark and Study for Long-horizon Agentic Programming in Command-Line Interfaces (2026.findings-acl)

Copied to clipboard

Yukang Feng, Jianwen Sun, Zelai Yang, Jiaxin Ai, Chuanhao Li, Zizhen Li, Fanrui Zhang, Kang He, Rui Ma, Jifan Lin, Jie Sun, Yang Xiao, Sizhuo Zhou, Wenxiao Wu, Yiming Liu, Pengfei Liu, Shenglin Zhang, Kaipeng Zhang

MoA: Heterogeneous Mixture of Adapters for Parameter-Efficient Fine-Tuning of Large Language Models (2026.acl-long)

Challenge:	Existing benchmarks for agentic programming in long-horizon command-line interface tasks are limited by short task horizons, data contamination from GitHub scraping, and a lack of fine-grained evaluation metrics.
Approach:	They propose a benchmark to evaluate agentic capabilities across long-horizon command-line interface tasks.
Outcome:	The proposed benchmarks cover four engineering categories: from scratch, feature addition, bug fixing, and refactoring.

Copied to clipboard

Jie Cao, Tianwei Lin, Bo Yuan, Rolan Yan, Hongyang He, Wenqiao Zhang, Juncheng Li, Dongping Zhang, Siliang Tang, Yueting Zhuang

Beyond Text: Incorporating Metadata and Label Structure for Multi-Label Document Classification using Heterogeneous Graphs (2021.emnlp-main)

Challenge:	Existing methods for parameter-efficient fine-tuning (PEFT) are limited by computational costs and performance degradation.
Approach:	They propose a method that integrates Low-Rank Adaptation and Mixture-of-Experts (MoE) they propose combining expert load imbalance and representation collapse to improve LLM performance .
Outcome:	The proposed method outperforms homogeneous MoE-LoRA architectures in performance and parameter efficiency.

Copied to clipboard

Chenchen Ye, Linhai Zhang, Yulan He, Deyu Zhou, Jie Wu

Information Extraction & Text Mining

Challenge:	Existing methods for multi-label document classification ignore the heterogeneous graphical structures of metadata and labels.
Approach:	They propose a neural network based approach to multi-label document classification that uses two heterogeneous graphs to model metadata and labels.
Outcome:	The proposed approach outperforms state-of-the-art models on two benchmark datasets.

On-Policy Self-Alignment with Fine-grained Knowledge Feedback for Hallucination Mitigation (2025.findings-acl)

Copied to clipboard

Xueru Wen, Jie Lou, Xinyu Lu, Yuqiu Ji, Xinyan Guan, Yaojie Lu, Hongyu Lin, Ben He, Xianpei Han, Debing Zhang, Le Sun

Reasoning

Challenge:	Large language models exhibit behavior that deviates from the boundaries of their knowledge during response generation.
Approach:	They propose a framework that allows large language models to explore their knowledge boundaries and self-correct generation behavior through fine-grained feedback signals.
Outcome:	The proposed framework enables LLMs to explore their knowledge boundaries and self-correct generation behavior through fine-grained feedback signals.

The Box is in the Pen: Evaluating Commonsense Reasoning in Neural Machine Translation (2020.findings-emnlp)

Copied to clipboard

Jie He, Tao Wang, Deyi Xiong, Qun Liu

Text Generation

Challenge:	a test suite to evaluate commonsense reasoning capability of neural machine translation is presented . language models pretrained on large-scale corpora achieve a commonsensing accuracy of lower than 72% on target translations of this test suite.
Approach:	They propose a test suite to evaluate the commonsense reasoning capability of neural machine translation.
Outcome:	The proposed test suite performs poorly on commonsense reasoning of the three ambiguity types in terms of reasoning accuracy and reasoning consistency.

Exploring Memorization in Fine-tuned Language Models (2024.acl-long)

Copied to clipboard

Shenglai Zeng, Yaxin Li, Jie Ren, Yiding Liu, Han Xu, Pengfei He, Yue Xing, Shuaiqiang Wang, Jiliang Tang, Dawei Yin

PsychEval: A Multi-Session and Multi-Therapy Benchmark for High-Realism AI Psychological Counselor (2026.findings-acl)

Challenge:	Existing studies have shown that pre-trained langauge models tend to memorize and regenerate segments of their pre-training corpus when prompted appropriately.
Approach:	They conduct the first comprehensive analysis to explore language models’ memorization during fine-tuning across tasks.
Outcome:	The proposed analysis shows that memorization presents a strong disparity among different fine-tuning tasks.

Copied to clipboard

Qianjun Pan, Junyi Wang, Jie Zhou, Yutao Yang, Junsong Li, Kaiyin Xu, Yougen Zhou, Yihan Li, JingYuan Zhao, Qin Chen, Ningning Zhou, Kai Chen, Liang He

Challenge:	Existing models focus on a single therapy, but complex cases require flexible strategies among various therapies.
Approach:	They propose a multi-session, multi-therapy, and highly realistic benchmark . it is designed to address three key challenges: 1) can we train a highly realistic AI counselor? 2) How to systematically evaluate an AI counselor?"
Outcome:	The proposed benchmark is annotated with extensive professional skills and includes over 677 meta-skills and 4577 atomic skills.

TamEdit: Trajectory-Aware Meta-Learning for Specificity-Preserving Continual Knowledge Editing (2026.acl-long)

Copied to clipboard

Shiqiang Tian, Cheng Ding, Qin Chen, Jie Zhou, Liang He

The Good and The Bad: Exploring Privacy Issues in Retrieval-Augmented Generation (RAG) (2024.findings-acl)

Challenge:	Existing methods for continual knowledge editing focus on single edits or preventing knowledge forgetting.
Approach:	They propose a meta-learning method that preserves specificity for continual knowledge editing by capturing relationships between different single edits within the trajectory.
Outcome:	Experiments show that TamEdit outperforms baselines in continual editing while preserving general capabilities.

Copied to clipboard

Shenglai Zeng, Jiankun Zhang, Pengfei He, Yiding Liu, Yue Xing, Han Xu, Jie Ren, Yi Chang, Shuaiqiang Wang, Dawei Yin, Jiliang Tang

Evaluating and Improving Graph to Text Generation with Large Language Models (2025.naacl-long)

Challenge:	Retrieval-augmented generation (RAG) is a powerful technique to facilitate language model generation with proprietary and private data, where data privacy is . a privacy issue that is currently under-explored, is posed by RAG.
Approach:	They propose to use retrieval-augmented generation (RAG) to facilitate language model generation with proprietary and private data where data privacy is a pivotal concern.
Outcome:	The proposed attack methods demonstrate that RAG can mitigate the old risks, i.e., leakage of the LLMs’ training data.

Copied to clipboard

Jie He, Yijun Yang, Wanqiu Long, Deyi Xiong, Victor Gutierrez Basulto, Jeff Z. Pan

Text Generation

Challenge:	Recent advances in large language models have revolutionized natural language processing due to their zero-and-short-shot capabilities.
Approach:	They propose a tuning-free prompting approach for graph-to-text generation tasks.
Outcome:	The proposed approach improves LLMs on graph-to-text generation tasks incrementally.

P-React: Synthesizing Topic-Adaptive Reactions of Personality Traits via Mixture of Specialized LoRA Experts (2025.findings-acl)

Copied to clipboard

Yuhao Dan, Jie Zhou, Qin Chen, Junfeng Tian, Liang He

Challenge:	Existing studies on personalized large language models focus on modeling explicit character profiles, while ignoring the underlying personality traits that truly shape behaviors and decision-making.
Approach:	They propose a personalized large language model (LLM) that captures implicit Big Five personality traits and integrates a Personality Specialization Loss to capture individual trait expressions.
Outcome:	The proposed model improves on Big Five personality traits and integrates a Personality Specialization Loss (PSL) to capture individual trait expressions.

A General Framework to Enhance Fine-tuning-based LLM Unlearning (2025.findings-acl)

Copied to clipboard

Jie Ren, Zhenwei Dai, Xianfeng Tang, Hui Liu, Jingying Zeng, Zhen Li, Rahul Goutam, Suhang Wang, Yue Xing, Qi He, Hui Liu

Learning Intrinsic Dimension via Information Bottleneck for Explainable Aspect-based Sentiment Analysis (2024.lrec-main)

Challenge:	Existing approaches to remove copyrighted and privacy-sensitive data from Large Language Models (LLMs) have been proposed to remove specific data from LLMs without requiring full retraining.
Approach:	They propose a general framework that enhances the utility of fine-tuning-based methods by distinguishing target data and suppressing related generations.
Outcome:	The proposed framework improves the unlearning and utility of fine-tuning-based methods by distinguishing the target data and suppressing related generations.

Copied to clipboard

Zhenxiao Cheng, Jie Zhou, Wen Wu, Qin Chen, Liang He

MiCEval: Unveiling Multimodal Chain of Thought’s Quality via Image Description and Reasoning Steps (2025.naacl-long)

Challenge:	Gradient-based explanation methods are increasingly used to interpret neural models in natural language processing (NLP) however, in the context of Aspect-based Sentiment Analysis, only specific dimensions are pertinent.
Approach:	They propose a Gradient-based explanation framework that leverages an information bottleneck to refine word embeddings into a concise intrinsic dimension, maintaining essential features and omitting unrelated information.
Outcome:	The proposed framework improves both the models’ performance and explanations’ clarity by identifying sentiment-aware features.

Copied to clipboard

Xiongtao Zhou, Jie He, Lanyu Chen, Jingyu Li, Haojing Chen, Victor Gutierrez Basulto, Jeff Z. Pan, Hanjie Chen

C-LLM: Learn to Check Chinese Spelling Errors Character by Character (2024.emnlp-main)

Challenge:	Existing methods for evaluating the quality of reasoning steps in multimodal chain-of-thought are lacking.
Approach:	They propose a framework to evaluate the correctness of reasoning chains by evaluating the quality of both the description and each reasoning step.
Outcome:	The proposed framework improves interpretability and human judgments on four state-of-the-art MLLMs.

Copied to clipboard

Kunting Li, Yong Hu, Liang He, Fandong Meng, Jie Zhou

Syntactic Text Processing

Challenge:	Chinese Spell Checking (CSC) aims to detect and correct spelling errors in sentences.
Approach:	They propose a Chinese Spell Checking method that learns to check errors Character by Character.
Outcome:	The proposed method achieves a 2.1% enhancement in general scenarios and a significant improvement in vertical domain scenarios compared to existing methods.

From Complex to Simple: Enhancing Multi-Constraint Complex Instruction Following Ability of Large Language Models (2024.findings-emnlp)

Copied to clipboard

Qianyu He, Jie Zeng, Qianxi He, Jiaqing Liang, Yanghua Xiao

Instances and Labels: Hierarchy-aware Joint Supervised Contrastive Learning for Hierarchical Multi-Label Text Classification (2023.findings-emnlp)

Challenge:	Large language models (LLMs) follow instructions with elaborate requirements, yet it remains under-explored how to enhance their ability to follow complex instructions with multiple constraints.
Approach:	They propose a method to obtain and utilize effective training data to enhance LLMs' ability to follow complex instructions with multiple constraints.
Outcome:	The proposed framework improves models' ability to follow instructions generally and generalize effectively across out-of-domain, in domain, and adversarial settings while maintaining general capabilities.

Copied to clipboard

Simon Chi Lok Yu, Jie He, Victor Basulto, Jeff Pan

Making Pre-trained Language Models Better Learn Few-Shot Spoken Language Understanding in More Practical Scenarios (2023.findings-acl)

Challenge:	Existing approaches to hierarchical multi-label text classification (HMTC) ignore the correlation between similar samples and introduce noise .
Approach:	They propose a semi-supervised method that uses a label hierarchy to bring text and label embeddings closer to each other by supervised contrastive learning.
Outcome:	The proposed method bridges the gap between supervised contrastive learning and HMTC by bringing text and label embeddings closer.

Copied to clipboard

Yufan Wang, Jie Mei, Bowei Zou, Rui Fan, Tingting He, Ai Ti Aw

OlympiadBench: A Challenging Benchmark for Promoting AGI with Olympiad-Level Bilingual Multimodal Scientific Problems (2024.acl-long)

Challenge:	Existing few-shot Spoken Language Understanding models need to be trained on a set of data-rich source domains and adapt to the target domain with a few examples.
Approach:	They propose a scenario where only a pre-trained language model and a few labeled examples are used to train few-shot SLU models.
Outcome:	The proposed model outperforms existing models on few-shot settings by reducing the number of slot labels and reducing training complexity.

Copied to clipboard

Chaoqun He, Renjie Luo, Yuzhuo Bai, Shengding Hu, Zhen Thai, Junhao Shen, Jinyi Hu, Xu Han, Yujie Huang, Yuxiang Zhang, Jie Liu, Lei Qi, Zhiyuan Liu, Maosong Sun

Scheduled Dialog Policy Learning: An Automatic Curriculum Learning Framework for Task-oriented Dialog System (2021.findings-acl)

Challenge:	Large Language Models (LLMs) and Large Multimodal Models have exceeded general human capabilities in various tasks.
Approach:	They present an Olympiad-level bilingual multimodal scientific benchmark featuring 8,476 problems from Olympiad level mathematics and physics competitions.
Outcome:	The best performing model, GPT-4V, attains an average score of 17.97% on OlympiadBench, with a mere 10.74% in physics, highlighting the benchmark rigor and the intricacy of physical reasoning.

Copied to clipboard

Sihong Liu, Jinchao Zhang, Keqing He, Weiran Xu, Jie Zhou

Dialogue Systems & Conversational Agents

Challenge:	et al., 2013) show that dialog policy learning is an important component of the task-oriented dialogue system.
Approach:	They propose a framework that integrates curriculum learning and policy optimization . they propose to train dialog agents from easy dialogues to complex ones .
Outcome:	The proposed framework outperforms the state-of-the-art model on multi-task dialogues.

Step-by-Step Mastery: Enhancing Soft Constraint Following Ability of Large Language Models (2025.findings-acl)

Copied to clipboard

Qingyu Ren, Jie Zeng, Qianyu He, Jiaqing Liang, Yanghua Xiao, Weikang Zhou, Zeye Sun, Fei Yu

SentiX: A Sentiment-Aware Pre-Trained Model for Cross-Domain Sentiment Analysis (2020.coling-main)

Challenge:	In real-world scenarios, user instructions often contain soft constraints, which are semantically related and cannot be rule-based verified, posing challenges for large language models.
Approach:	They propose a pipeline to construct datasets with high-quality outputs for instructions containing soft constraints automatically and use Direct Preference Optimization (DPO) as the training method.
Outcome:	The proposed model improves the LLMs' soft constraint following ability by using direct preference optimization (DPO) and constraint quantity.

Copied to clipboard

Jie Zhou, Junfeng Tian, Rui Wang, Yuanbin Wu, Wenming Xiao, Liang He

On the Generalization of Training-based ChatGPT Detection Methods (2024.findings-emnlp)

Challenge:	Pre-trained language models have been widely applied to cross-domain NLP tasks like sentiment analysis, but fine-tuning them on the source domain tends to overfit, leading to inferior results on the target domain.
Approach:	They propose to pre-train a sentiment-aware language model (SentiX) via domain-invariant sentiment knowledge from large-scale review datasets and utilize it for cross-domain sentiment analysis tasks without fine-tuning.
Outcome:	The proposed model achieves state-of-the-art in all the cross-domain sentiment analysis tasks and can be trained with only 1% samples and better than BERT with 90% samples.

Copied to clipboard

Han Xu, Jie Ren, Pengfei He, Shenglai Zeng, Yingqian Cui, Amy Liu, Hui Liu, Jiliang Tang

Prototype-based HyperAdapter for Sample-Efficient Multi-task Tuning (2023.emnlp-main)

Challenge:	Existing studies show that training-based methods are ineffective to detect LLM generated texts from unseen tasks or topics which are not collected during training.
Approach:	They propose to train classification models to distinguish LLMs from human texts by a distribution shift caused by prompts, text lengths, topics, and language tasks.
Outcome:	The proposed methods can detect LLMs from black-box models, but they suffer from distribution shifts due to a wide range of factors, including prompts, text lengths, topics, and language tasks.

Copied to clipboard

Hao Zhao, Jie Fu, Zhaofeng He

SCAIR: Schema-Conditioned Agentic Iterative Reasoning for Enterprise Knowledge Graphs (2026.acl-industry)

Challenge:	Existing methods for fine-tuning pre-trained language models are limited to low-data regimes and require learning different modules to adapt to diverse tasks.
Approach:	They propose a framework for parameter-efficient fine-tuning that trains modules per task . they use an instance-dense retriever and a prototypical hypernetwork to generate conditional modules .
Outcome:	The proposed framework outperforms existing methods on multi-task learning and few-shot transfer learning.

Copied to clipboard

Prateek Chaturvedi, Yuqicheng Zhu, Hongkuan Zhou, Dongzhuoran Zhou, Yunjie He, Steffen Staab, Fei Du, Jie Tang, Evgeny Kharlamov

Structured Data in NLP

Challenge:	Existing agentic approaches for Knowledge Graph-based Retrieval-Augmented Generation fail to generalize to real-world enterprise Knowledge graphs (KGs) dense, schema-driven, and operationally constrained, requiring a training-free framework.
Approach:	They propose a training-free framework that integrates structured planning with controlled iterative reasoning by injecting schema-conditioned structural priors and enforcing schemas during multi-hop reasoning.
Outcome:	The proposed framework significantly improves on a real-world enterprise-oriented benchmark constructed from a Configuration Management DataBase (CMDB).

Instructions are all you need: Self-supervised Reinforcement Learning for Instruction Following (2026.acl-long)

Copied to clipboard

Qingyu Ren, Qianyu He, Powei Chang, Jie Zeng, Zeye Sun, Fei Yu, Jiaqing Liang, Yanghua Xiao

UltraEval: A Lightweight Platform for Flexible and Comprehensive Evaluation for LLMs (2024.acl-demos)

Challenge:	Existing reinforcement learning approaches suffer from dependency on external supervision and sparse reward signals from multi-constraint tasks.
Approach:	They propose a self-supervised reinforcement learning framework that eliminates dependency on external supervision by deriving reward signals directly from instructions and generating pseudo-labels for reward model training.
Outcome:	The proposed framework achieves strong improvements across 3 in-domain and 5 out-of-domain datasets while maintaining computational efficiency.

Copied to clipboard

Chaoqun He, Renjie Luo, Shengding Hu, Ranchi Zhao, Jie Zhou, Hanghao Wu, Jiajie Zhang, Xu Han, Zhiyuan Liu, Maosong Sun

A Survey of Reinforcement Learning for Large Language Models under Data Scarcity: Challenges and Solutions (2026.acl-long)

Challenge:	Existing evaluation platforms are complex and poorly modularized, hindering seamless incorporation into researcher’s workflows.
Approach:	They propose a lightweight evaluation framework characterized by lightweight, comprehensiveness, modularity, and efficiency that integrates models, data, and metrics into a unified evaluation workflow.
Outcome:	The proposed evaluation framework is lightweight, comprehensive, modular, and efficient.

Copied to clipboard

Zhiyin Yu, Yuchen Mou, Juncheng Yan, Junyu Luo, Chunchun Chen, Xing Wei, Yunhui Liu, Hongru Sun, Yuxing Zhang, Jun Xu, Yatao Bian, Ming Zhang, Wei Ye, Tieke He, Jie Yang, Guanjie Zheng, Zhonghai Wu, Bo Zhang, Lei Bai, Xiao Luo

Mitigating the Privacy Issues in Retrieval-Augmented Generation (RAG) via Pure Synthetic Data (2025.emnlp-main)

Challenge:	Existing research on reinforcement learning for LLMs under data scarcity has not been unified.
Approach:	They propose a top-up hierarchical framework built around three complementary perspectives: data-centric, training-centric and framework-centric.
Outcome:	The proposed framework provides a clear conceptual foundation for understanding the design space of data-efficient RL for large language models and to guide researchers working in this emerging area.

Copied to clipboard

Shenglai Zeng, Jiankun Zhang, Pengfei He, Jie Ren, Tianqi Zheng, Hanqing Lu, Han Xu, Hui Liu, Yue Xing, Jiliang Tang

Attending via both Fine-tuning and Compressing (2021.findings-acl)

Challenge:	Existing literature suggests that RAG systems may face privacy issues when the retrieval process involves private data.
Approach:	They propose a two-stage synthetic data generation paradigm that uses attributes to preserve contextual information from the original data.
Outcome:	The proposed approach preserves key contextual information from the original data while reducing privacy risks.

Copied to clipboard

Jie Zhou, Yuanbin Wu, Qin Chen, Xuanjing Huang, Liang He

Explainability & Interpretability in NLP

Challenge:	Existing studies show that attention mechanisms can improve models' interpretation, but they are not explicable.
Approach:	They propose a framework consisting of a learner and a compressor to purify attention scores . they propose to fine-tune and compress the attention mechanism to obtain a more faithful explanation .
Outcome:	The proposed framework improves performance and interpretability on eight benchmark datasets.

LongAlign: A Recipe for Long Context Alignment of Large Language Models (2024.findings-emnlp)

Copied to clipboard

Yushi Bai, Xin Lv, Jiajie Zhang, Yuze He, Ji Qi, Lei Hou, Jie Tang, Yuxiao Dong, Juanzi Li

TGEA: An Error-Annotated Dataset and Benchmark Tasks for TextGeneration from Pretrained Language Models (2021.acl-long)

Challenge:	Existing studies to build long context language models focus on context extension and continual training on long text.
Approach:	They propose a recipe for instruction fine-tuning on input sequences of similar length . they adopt packing and sorted batching strategies to speed up supervised fine-uning .
Outcome:	The proposed model outperforms existing recipes for LLMs in long context tasks by 30% while maintaining proficiency in handling short, generic tasks.

Copied to clipboard

Jie He, Bo Peng, Yi Liao, Qun Liu, Deyi Xiong

A Survey of Inductive Reasoning for Large Language Models (2026.acl-long)

Challenge:	Using pretrained language models, we propose an error-annotated dataset for text generation . we use carefully selected prompt words to guide GPT-2 to generate candidate sentences .
Approach:	They propose an error-annotated dataset with multiple benchmark tasks for text generation from pretrained language models.
Outcome:	The proposed dataset covers 24 types of errors according to common sense and linguistics.

Copied to clipboard

Kedi Chen, Dezhao Ruan, Yuhao Dan, Yaoting Wang, Siyu Yan, Xuecheng Wu, Yinqi Zhang, Qin Chen, Jie Zhou, Liang He, Biqing Qi, Linyang Li, Qipeng Guo, Xiaoming Shi, Wei Zhang

An Information Minimization Based Contrastive Learning Model for Unsupervised Sentence Embeddings Learning (2022.coling-1)

Challenge:	Inductive reasoning is an important task for large language models (LLMs).
Approach:	They propose a survey of inductive reasoning for large language models . they categorize methods into three main areas: post-training enhancement, test-time exploration, and data augmentation.
Outcome:	The proposed method improves inductive reasoning in large language models.

Copied to clipboard

Shaobin Chen, Jie Zhou, Yuling Sun, Liang He

Cheems: A Practical Guidance for Building and Evaluating Chinese Reward Models from Scratch (2025.acl-long)

Challenge:	Recent contrastive learning methods keep positive pairs similar and push negative pairs apart, which leads to redundant information in sentence embeddings.
Approach:	They propose a contrastive learning approach which maximizes mutual information and minimizes the information entropy between positive and negative instances.
Outcome:	The proposed model outperforms all previous competitors on supervised and unsupervised tasks.

Copied to clipboard

Xueru Wen, Jie Lou, Zichao Li, Yaojie Lu, XingYu XingYu, Yuqiu Ji, Guohai Xu, Hongyu Lin, Ben He, Xianpei Han, Le Sun, Debing Zhang

Challenge:	Existing Chinese resources are small in scale and limited to specific domains, making them insufficient for LLM post-training.
Approach:	They propose a Chinese-annotated reward model and a preference dataset to address this gap . they evaluate Chinese RMs on CheemsBench and construct an RM that captures human preferences .
Outcome:	The proposed RM achieves state-of-the-art performance on CheemsBench and CheeMePreference.

Is “hot pizza” Positive or Negative? Mining Target-aware Sentiment Lexicons (2021.eacl-main)

Copied to clipboard

Jie Zhou, Yuanbin Wu, Changzhi Sun, Liang He

Sentiment Analysis

Challenge:	Existing sentiment lexicons assume words’ sentiments are invariant within a domain, but this assumption is weak for fine-granularity analyses of text sentiments.
Approach:	They propose a "perturb-and-see" method to extract commonsense sentiments from large-scale datasets by binding a word's sentiment to its collocation words instead of domain labels.
Outcome:	The proposed framework is able to achieve highly competitive performances on the unsupervised opinion relation extraction task.

Order Matters: Investigate the Position Bias in Multi-constraint Instruction Following (2025.findings-acl)

Copied to clipboard

Jie Zeng, Qianyu He, Qingyu Ren, Jiaqing Liang, Weikang Zhou, Zeye Sun, Fei Yu, Yanghua Xiao

Challenge:	Existing large language models struggle to follow multi-constraint instructions in real-world applications.
Approach:	They propose to quantify the difficulty distribution of constraints by a novel Difficulty Distribution Index (CDDI) they find that LLMs are more performant when presented with constraints in a “hard-to-easy” order.
Outcome:	The proposed model is more performant when presented with constraints in a “hard-to-easy” order, compared with existing models with different architectures and sizes of parameters.

MINTQA: A Multi-Hop Question Answering Benchmark for Evaluating LLMs on New and Long-tail Knowledge (2026.acl-long)

Copied to clipboard

Jie He, Nan Hu, Wanqiu Long, Jiaoyan Chen, Jeff Z. Pan

Natural Language Interfaces

Challenge:	Existing studies have shown that large language models can handle knowledge with varying familiarity.
Approach:	They propose a benchmark to evaluate multi-hop question answering on new and tail knowledge . they use RAG to integrate external knowledge into large language models .
Outcome:	The proposed benchmark evaluates the multi-hop reasoning ability of large language models . it primarily evaluates their ability to handle knowledge with different levels of familiarity .

MT-Bench-101: A Fine-Grained Benchmark for Evaluating Large Language Models in Multi-Turn Dialogues (2024.acl-long)

Copied to clipboard

Ge Bai, Jie Liu, Xingyuan Bu, Yancheng He, Jiaheng Liu, Zhanhui Zhou, Zhuoran Lin, Wenbo Su, Tiezheng Ge, Bo Zheng, Wanli Ouyang

Natural Language Interfaces

Challenge:	Large Language Models (LLMs) have greatly enhanced dialogue systems, but evaluation of their capabilities remains a challenge.
Approach:	They propose a model to evaluate the fine-grained abilities of Large Language Models in multi-turn dialogues.
Outcome:	The proposed model evaluates 21 popular chatbots based on MT-Bench-101 . it includes 3 overarching abilities and 13 distinct tasks within multi-turn dialogue scenarios.

GraphReader: Building Graph-based Agent to Enhance Long-Context Abilities of Large Language Models (2024.findings-emnlp)

Copied to clipboard

Shilong Li, Yancheng He, Hangyu Guo, Xingyuan Bu, Ge Bai, Jie Liu, Jiaheng Liu, Xingwei Qu, Yangguang Li, Wanli Ouyang, Wenbo Su, Bo Zheng

Enhancing One-Shot Pruned Pre-trained Language Models through Sparse-Dense-Sparse Mechanism (2025.coling-main)

Challenge:	Existing models for long contexts struggle to handle long inputs due to limited context window and memory usage.
Approach:	They propose a graph-based agent system that analyzes long texts into a graphical graph . GraphReader consistently outperforms GPT-4-128k across context lengths from 16k to 256k .
Outcome:	The proposed model outperforms existing models on four challenging benchmarks.

Copied to clipboard

Guanchen Li, Xiandong Zhao, Lian Liu, Zeping Li, Yixing Xu, Dong Li, Lu Tian, Jie He, Ashish Sirasao, Emad Barsoum

An Empirical Study on Parameter-Efficient Fine-Tuning for MultiModal Large Language Models (2024.findings-acl)

Challenge:	Pre-trained language models (PLMs) are robust in contextual understanding but their considerable size incurs significant computational and storage costs.
Approach:	They propose a Sparse-Dense-Sparse pruning framework to prune PLMs . they prune less critical connections using conventional pruning methods .
Outcome:	The proposed pruning framework outperforms SparseGPT and Wanda under identical sparsity.

Copied to clipboard

Xiongtao Zhou, Jie He, Yuhua Ke, Guangyao Zhu, Victor Gutierrez Basulto, Jeff Pan

LLM-KT: Enhancing Large Language Models with Knowledge Tracing via Multi-Level Plug-and-Play Alignment (2026.findings-acl)

Challenge:	Multimodal Large Language Models fine-tuned with multimodal instruction-following data have demonstrated formidable capabilities in multimodal tasks.
Approach:	They propose to employ four PEFT methods to fine-tune the LLM component of open-source MLLMs.
Outcome:	The proposed method is the best performing on seven datasets, while fine-tuning the connector layers leads to improved performance in most MLLMs.

Copied to clipboard

Ziwei Wang, Jie Zhou, Qin Chen, Bo Jiang, Qingchun Bai, Liang Dou, Liang He

A.S.E: A Repository-Level Benchmark for Evaluating Security in AI-Generated Code (2026.findings-acl)

Challenge:	Existing methods to learn behavioral sequences fail to capture complex behavioral patterns due to a lack of deep reasoning capabilities and world knowledge.
Approach:	They propose a framework that integrates the reasoning power of Large Language Models with the sequential modeling strengths of traditional KT methods via multi-level plug-and-play alignment.
Outcome:	Extensive experiments on four standard datasets show that the proposed framework outperforms existing methods on state-of-the-art questions.

Copied to clipboard

Keke Lian, Wang Bin, Lei Zhang, Libo Chen, Junjie Wang, Ziming Zhao, Yujiu Yang, Miaoqian Lin, Haotong Duan, Haoran Zhao, Shuang Liao, Mingda Guo, Quan Jiazheng, Yilu Zhong, Chenhao He, Chen Zichuan, Jie Wu, Haoling Li, Zhaoxuan Li, Jiongchi Yu, Hui LI, Dong Zhang

Programming Languages in NLP

Challenge:	Existing security evaluation benchmarks lack relevance to real-world AI programming tasks . current LLMs struggle with secure coding, research shows .
Approach:	They propose a repository-level evaluation benchmark to assess security of AI-generated code.
Outcome:	The proposed framework mirrors real-world AI programming tasks and offers valuable insights into the state of AI code generation.

GenTool: Enhancing Tool Generalization in Language Models through Zero-to-One and Weak-to-Strong Simulation (2025.findings-acl)

Copied to clipboard

Jie He, Jennifer Neville, Mengting Wan, Longqi Yang, Hui Liu, Xiaofeng Xu, Xia Song, Jeff Z. Pan, Pei Zhou