Papers by Xianpei Han

ARise: Towards Knowledge-Augmented Reasoning via Risk-Adaptive Search (2025.acl-long)

Challenge:	Existing code sandboxes fail to provide accurate verification and efficiency under high-concurrency workloads.
Approach:	They propose a high-fidelity code verification system that provides sandbox feedback for RL training and evaluation.
Outcome:	The proposed system outperforms heuristic-matching baselines on LiveCodeBench and training stability on high-concurrency workloads.

Copied to clipboard

Yize Zhang, Tianshu Wang, Sirui Chen, Kun Wang, Xingyu Zeng, Hongyu Lin, Xianpei Han, Le Sun, Chaochao Lu

All Languages Matter: Understanding and Mitigating Language Bias in Multilingual RAG (2026.acl-long)

Challenge:	Large language models (LLMs) have impressive capabilities but their application in open-ended, knowledge-intensive, complex reasoning scenarios is limited.
Approach:	They propose a framework that integrates risk assessment of intermediate reasoning states with dynamic retrieval-augmented generation within a Monte Carlo tree search paradigm.
Outcome:	The proposed framework outperforms the state-of-the-art KAR methods by up to 23.10% and the latest RAG-equipped large reasoning models by upto 25.37%.

Copied to clipboard

Dan Wang, Guozhao Mo, Yafei Shi, Cheng Zhang, Bo Zheng, Boxi Cao, Xuanang Chen, Yaojie Lu, Hongyu Lin, Ben He, Xianpei Han, Le Sun

Multilinguality

Challenge:	Existing mRAG systems suffer from a language bias during reranking, systematically favoring English and the query’s native language.
Approach:	They propose a language-agnostic utility-driven reranker alignment technique to mitigate language bias during re-ranking.
Outcome:	The proposed approach mitigates language bias and consistently improves mRAG performance across languages.

Learning In-context Learning for Named Entity Recognition (2023.acl-long)

Copied to clipboard

Jiawei Chen, Yaojie Lu, Hongyu Lin, Jie Lou, Wei Jia, Dai Dai, Hua Wu, Boxi Cao, Xianpei Han, Le Sun

Teach Small Models to Reason by Curriculum Distillation (2025.emnlp-main)

Challenge:	Existing methods to recognize entities in text are limited by the diversity of entity types and the lack of high-quality annotations.
Approach:	They propose an in-context learning-based NER approach that can inject in-const NER ability into PLMs and recognize entities of novel types on-the-fly using only a few demonstrative instances.
Outcome:	The proposed method outperforms the PLMs+fine-tuning counterparts on 4 few-shot NER datasets and significantly outperformed the Plms+initialized extractors.

Copied to clipboard

Wangyi Jiang, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun

Distilling Discrimination and Generalization Knowledge for Event Detection via Delta-Representation Learning (P19-1)

Challenge:	Large Reasoning Models (LRMs) show strong System-2-style reasoning, but at the cost of significant computational overhead.
Approach:	They propose a two-stage curriculum distillation framework which builds a robust internal problem-solving student model and then teaches the student model to externalize this knowledge as explicit reasoning.
Outcome:	The proposed model outperforms single-stage baselines on mathematical benchmarks and significantly outperformed LRMs on complex tasks.

Copied to clipboard

Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun

Learning or Self-aligning? Rethinking Instruction Fine-tuning (2024.acl-long)

Challenge:	Current neural event detection approaches focus on trigger-centric representations, which work well on distilling discrimination knowledge, but poorly on learning generalization knowledge.
Approach:	They propose a Delta-learning approach to distill discrimination and generalization knowledge by incrementally learning and adaptively fusing event representation.
Outcome:	The proposed method significantly outperforms previous approaches on unseen/sparse trigger words and achieves state-of-the-art performance on ACE2005 and KBP2017 datasets.

Copied to clipboard

Mengjie Ren, Boxi Cao, Hongyu Lin, Cao Liu, Xianpei Han, Ke Zeng, Wan Guanglu, Xunliang Cai, Le Sun

REInstruct: Building Instruction Data from Unlabeled Corpus (2024.findings-acl)

Challenge:	Instruction fine-tuning (IFT) is a crucial phase in building large language models (LLMs).
Approach:	They propose a knowledge intervention framework to decouple the potential underlying factors of IFT and enable individual analysis of different factors.
Outcome:	The proposed framework decouples the potential underlying factors of IFT, enabling individual analysis of different factors.

Copied to clipboard

Shu Chen, Xinyan Guan, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun

MemSearcher: Iterative Memory Integration for Search Agent via End-to-End Reinforcement Learning (2026.findings-acl)

Challenge:	Existing methods for annotating instruction data are expensive and difficult to scale.
Approach:	They propose a method to automatically build instruction data from an unlabeled corpus without heavy reliance on proprietary LLMs and human annotation.
Outcome:	The proposed method outperforms existing methods on AlpacaEval leaderboard and other open-source methods.

Copied to clipboard

Qianhao Yuan, Jie Lou, Zichao Li, Jiawei Chen, Yaojie Lu, Hongyu Lin, Le Sun, Debing Zhang, Xianpei Han

Executing Natural Language-Described Algorithms with Large Language Models: An Investigation (2024.lrec-main)

Challenge:	Recent LLM-based search agents often concatenate the full interaction history into the context, producing long and noisy inputs and increasing compute cost and memory overhead.
Approach:	They propose an agent framework that maintains a compact memory during multi-turn interactions.
Outcome:	The proposed framework outperforms strong history-concatenation (ReAct-style) baselines on a range of public datasets while maintaining nearly constant token counts across multi-turn interactions.

Copied to clipboard

Xin Zheng, Qiming Zhu, Hongyu Lin, Yaojie Lu, Xianpei Han, Le Sun

Unified Structure Generation for Universal Information Extraction (2022.acl-long)

Challenge:	Recent advances in large language models (LLMs) have revolutionized the field of natural language processing and artificial intelligence, creating new SOTAs and reaching human-level language understanding performance on a series of tasks and benchmarks.
Approach:	They propose to use an algorithm test set sourced from Introduction to Algorithm to assess LLMs' code execution abilities.
Outcome:	The proposed model can execute programs described in natural language as long as no heavy numeric computation is involved.

Copied to clipboard

Yaojie Lu, Qing Liu, Dai Dai, Xinyan Xiao, Hongyu Lin, Xianpei Han, Le Sun, Hua Wu

DeepPresenter: Environment-Grounded Reflection for Agentic Presentation Generation (2026.findings-acl)

Challenge:	Information extraction suffers from its varying targets, heterogeneous structures, and demand-specific schemas.
Approach:	They propose a unified text-to-structure generation framework, namely UIE, which can universally model different IE tasks, adaptively generate targeted structures, and collaboratively learn general IE abilities from different knowledge sources.
Outcome:	The proposed framework can model different IE tasks, generate targeted structures, and learn general IE abilities from different knowledge sources.

Copied to clipboard

Hao Zheng, Guozhao Mo, Xinru Yan, Qianhao Yuan, Wenkai Zhang, Xuanang Chen, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun

Challenge:	Existing presentation agents rely on predefined workflows and fixed templates to generate presentations.
Approach:	They propose an agentic framework that adapts to diverse user intents and iterative refinement based on observation.
Outcome:	The proposed framework can be used to generate presentations with environmental observations.

Few-shot Named Entity Recognition via Superposition Concept Discrimination (2024.lrec-main)

Copied to clipboard

Jiawei Chen, Hongyu Lin, Xianpei Han, Yaojie Lu, Shanshan Jiang, Bin Dong, Le Sun

PaperRegister: Boosting Flexible-grained Paper Search via Hierarchical Register Indexing (2026.acl-long)

Challenge:	Few-shot named entity recognition (NER) aims to identify entities of target types with limited number of illustrative instances.
Approach:	They propose a superposition concept discriminator which solves the intrinsic generalization problem by an active learning paradigm.
Outcome:	The proposed model significantly improves few-shot named entity recognition (FS-NER) with minimal additional efforts.

Copied to clipboard

Zhuoqun Li, Xuanang Chen, Hongyu Lin, Yaojie Lu, Xianpei Han, Shanshan Jiang, Bin Dong, Le Sun

From Paraphrasing to Semantic Parsing: Unsupervised Semantic Parsing via Synchronous Semantic Decoding (2021.acl-long)

Challenge:	Existing paper search systems lack detailed information to support finer-grained queries.
Approach:	They propose a paper-based index that transforms abstract-based corpus index into hierarchical index tree and offline can support paper search queries.
Outcome:	The proposed system achieves the SOTA performance and excels in fine-grained scenarios.

Copied to clipboard

Shan Wu, Bo Chen, Chunlei Xin, Xianpei Han, Le Sun, Weipeng Zhang, Jiansong Chen, Fan Yang, Xunliang Cai

Adaptive Scaling for Sparse Detection in Information Extraction (P18-1)

Challenge:	Experimental results show that Synchronous Semantic Decoding (SSD) can achieve state-of-the-art unsupervised semantic parsing performance on multiple datasets.
Approach:	They propose an unsupervised method which solves the semantic gap and the structure gap by leveraging paraphrasing and grammar-constrained decoding.
Outcome:	The proposed method can solve the semantic gap and structure gap on multiple datasets.

Copied to clipboard

Semantic-aware Contrastive Learning for More Accurate Semantic Parsing (2022.emnlp-main)

Challenge:	Detection problems involving positive instances are often deficient in information extraction tasks . a number of researches have employed neural network models to solve detection problems .
Approach:	They propose an algorithm which can handle positive sparsity problem and directly optimize over F-measure . they borrow the idea of marginal utility from economics and propose a theoretical framework for instance importance measuring .
Outcome:	The proposed algorithm improves on positive sparsity problem and over F-measure . it leads to more effective and stable training of neural network based detection models.

Copied to clipboard

Shan Wu, Chunlei Xin, Bo Chen, Xianpei Han, Le Sun

Sequence-to-Action: End-to-End Semantic Graph Generation for Semantic Parsing (P18-1)

Challenge:	Existing studies on semantic parsing use Maximum Likelihood Estimation (MLE) to train discriminative semantic parses.
Approach:	They propose a semantic-aware contrastive learning algorithm which can learn to distinguish fine-grained meaning representations and take the overall sequence-level semantic into consideration.
Outcome:	The proposed algorithm improves on two standard datasets and gets state-of-the-art performance over existing methods.

Copied to clipboard

Bo Chen, Le Sun, Xianpei Han

Navigating the Infinite Dynamic Web Space: Effective In-Context Exploration via Cognitive Multi-Agent Collaboration (2026.eacl-long)

Challenge:	Existing methods for semantic parsing are difficult to design and learn, especially in wideopen domains.
Approach:	They propose a neural semantic parsing approach which models semantic par- sing as an end-to-end semantic graph generation process.
Outcome:	The proposed model achieves state-of-the-art performance on Overnight dataset and gets competitive performance on Geo and Atis datasets.

Copied to clipboard

Guozhao Mo, Yanjiang Liu, Yafei Shi, Jiawei Chen, Yang Li, Yaojie Lu, Hongyu Lin, Ben He, Le Sun, Bo Zheng, Xianpei Han

Linguistics & Cognitive NLP

Challenge:	Existing methods for dynamic web navigation rely on greedy strategies or value estimation, struggle to achieve effective backtracking and are heavily dependent on proprietary models.
Approach:	They propose a cognitive multi-agent collaboration framework that enhances cyberspace exploration capability through In-Context Exploration.
Outcome:	The proposed framework surpasses the proprietary model Claude-3.5 Sonnet on the WebArena benchmark.

Memorizing is Not Enough: Deep Knowledge Injection Through Reasoning (2025.acl-long)

Copied to clipboard

Ruoxi Xu, Yunjie Ji, Boxi Cao, Yaojie Lu, Hongyu Lin, Xianpei Han, Ben He, Yingfei Sun, Xiangang Li, Le Sun

A Rigorous Study on Named Entity Recognition: Can Fine-tuning Pretrained Model Lead to the Promised Land? (2020.emnlp-main)

Challenge:	Existing knowledge injection frameworks focus on knowledge memorization and retrieval, but static nature of large language models leads to outdated information as the real world evolves or when adapting to domain-specific knowledge.
Approach:	They propose a four-tier knowledge injection framework that defines the levels of knowledge injection: memorization, retrieval, reasoning, and association.
Outcome:	The proposed framework defines the levels of knowledge injection: memorization, retrieval, reasoning, and association.

Copied to clipboard

Hongyu Lin, Yaojie Lu, Jialong Tang, Xianpei Han, Le Sun, Zhicheng Wei, Nicholas Jing Yuan

Few-shot Named Entity Recognition with Self-describing Networks (2022.acl-long)

Challenge:	Named entity recognition (NER) is a fundamental task of information extraction.
Approach:	They propose to perform randomization tests on standard NER benchmarks to examine name regularity, mention coverage and context diversity.
Outcome:	The proposed model performs better on standard NER benchmarks than other models on open datasets.

Copied to clipboard

Jiawei Chen, Qing Liu, Hongyu Lin, Xianpei Han, Le Sun

ConsistentChat: Building Skeleton-Guided Consistent Multi-Turn Dialogues for Large Language Models from Scratch (2025.emnlp-main)

Challenge:	Existing few-shot named entity recognition (NER) models capture information from limited instances while transferring useful knowledge from external resources.
Approach:	They propose a self-describing mechanism for few-shot NER which can universally describe mentions using concepts and automatically map novel entity types to concepts.
Outcome:	The proposed model can universally describe mentions using concepts and automatically map novel entity types to concepts and adaptively recognize entities on-demand.

Copied to clipboard

Jiawei Chen, Xinyan Guan, Qianhao Yuan, Mo Guozhao, Weixiang Zhou, Yaojie Lu, Hongyu Lin, Ben He, Le Sun, Xianpei Han

Dialogue Systems & Conversational Agents

Challenge:	Existing instruction data synthesis methods focus on single-turn instructions and neglect cross-turn coherence, resulting in context drift and reduced task completion rates.
Approach:	They propose a framework that constrains multi-turn instruction synthesis by explicitly modeling human conversational intent.
Outcome:	The proposed framework outperforms existing models trained on single-turn and multi-turn instruction datasets.

Critic-CoT: Boosting the Reasoning Abilities of Large Language Model via Chain-of-Thought Critic (2025.findings-acl)

Copied to clipboard

Xin Zheng, Jie Lou, Boxi Cao, Xueru Wen, Yuqiu Ji, Hongyu Lin, Yaojie Lu, Xianpei Han, Debing Zhang, Le Sun

ChatGPT Is a Knowledgeable but Inexperienced Solver: An Investigation of Commonsense Problem in Large Language Models (2024.lrec-main)

Challenge:	Existing approaches to improve the reasoning performance of large language models rely on intuitive instance-level feedback, which limits the reasoning capabilities.
Approach:	They propose a framework that pushes LLMs toward System-2-like critic capability by using a step-wise CoT reasoning paradigm and automatic construction of weak-supervision data without human annotation.
Outcome:	The proposed model significantly improves task-solving performance by filtering out invalid solutions or iterative refinement.

Copied to clipboard

Ning Bian, Xianpei Han, Le Sun, Hongyu Lin, Yaojie Lu, Ben He, Shanshan Jiang, Bin Dong

Natural Language Interfaces

Challenge:	acquiring and representing commonsense in machines has posed a long-standing challenge (Li et al., 2021; Zhang e t al, 2022; Zhou e al. 2023) .
Approach:	They use a commonsense-based LLM to evaluate ChatGPT's commonsensing abilities by analyzing 11 datasets and generating knowledge descriptions.
Outcome:	The proposed model can achieve good QA accuracies while still struggling with certain domains of datasets.

RMTBench: Benchmarking LLMs Through Multi-Turn User-Centric Role-Playing (2025.findings-emnlp)

Copied to clipboard

Hao Xiang, Tianyi Tang, Yang Su, Bowen Yu, An Yang, Fei Huang, Yichang Zhang, Yaojie Lu, Hongyu Lin, Xianpei Han, Jingren Zhou, Junyang Lin, Le Sun

Dialogue Systems & Conversational Agents

Challenge:	Existing benchmarks focus on character-centric approach and fail to reflect real-world applications.
Approach:	RMTBench is a user-centric bilingual role-playing benchmark featuring 80 diverse characters and over 8,000 dialogue rounds.
Outcome:	RMTBench features 80 diverse characters and over 8,000 dialogue rounds.

Knowledgeable or Educated Guess? Revisiting Language Models as Knowledge Bases (2021.acl-long)

Copied to clipboard

Boxi Cao, Hongyu Lin, Xianpei Han, Le Sun, Lingyong Yan, Meng Liao, Tong Xue, Jin Xu

Honey or Poison? Solving the Trigger Curse in Few-shot Event Detection via Causal Intervention (2021.emnlp-main)

Challenge:	Recent studies show that pre-trained masked language models can be factual knowledge bases.
Approach:	They conduct a rigorous study to explore the underlying predicting mechanisms of MLMs . they find that previous decent performance mainly owes to the biased prompts which overfit dataset artifacts a .
Outcome:	The proposed model improves on illustrative cases and external contexts . the results question the previous findings that MLMs can be reliable factual knowledge bases .

Copied to clipboard

Jiawei Chen, Hongyu Lin, Xianpei Han, Le Sun

Retentive or Forgetful? Diving into the Knowledge Memorizing Mechanism of Language Models (2024.lrec-main)

Challenge:	Recent supervised ED approaches have achieved promising performance but require large number of manually annotated event data.
Approach:	They propose to overfit the trigger confounder of the context and the result . they propose to intervene on the context via backdoor adjustment during training .
Outcome:	The proposed method significantly improves the FSED on ACE05 and MAVEN datasets.

Copied to clipboard

Boxi Cao, Qiaoyu Tang, Hongyu Lin, Shanshan Jiang, Bin Dong, Xianpei Han, Jiawei Chen, Tianshu Wang, Le Sun

Not All Contexts Are Equal: Teaching LLMs Credibility-aware Generation (2024.emnlp-main)

Challenge:	Pre-trained language models have shown remarkable memory formation, but vanilla networks without pre-training suffer catastrophic forgetting problem.
Approach:	They conduct experiments to investigate the retentive-forgetful contradiction between vanilla and pre-trained language models by controlling the target knowledge types, learning strategies and learning schedules.
Outcome:	The results show that pre-trained language models are forgetful and pre-training leads to retentive models .

Copied to clipboard

Ruotong Pan, Boxi Cao, Hongyu Lin, Xianpei Han, Jia Zheng, Sirui Wang, Xunliang Cai, Le Sun

Across Programming Language Silos: A Study on Cross-Lingual Retrieval-Augmented Code Generation (2026.findings-acl)

Challenge:	Existing RAG paradigms suffer from the impact of flawed information introduced during the retrieval phrase, thereby diminishing the reliability and correctness of the generated output.
Approach:	They propose a framework that empowers models to discern and process information based on its credibility.
Outcome:	The proposed framework outperforms existing models with retrieval augmentation and exhibits robustness despite increasing noise in the context.

Copied to clipboard

Qiming Zhu, Jialun Cao, Xuanang Chen, Weili Zhang, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun, Shing-Chi Cheung

Programming Languages in NLP

Challenge:	Current research on large language models with retrieval-augmented code generation (RACG) has focused on single-language settings, leaving their cross-lingual effectiveness underexplored.
Approach:	They construct a dataset covering 13 PLs with nearly 14K instances to study cross-lingual code knowledge transfer in RACG.
Outcome:	The proposed model shows unequal cross-lingual knowledge transfer even with direct injection and shows limited reliance on natural language information embedded in code when equipped with a code-specific retriever.

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

Semi-Supervised Lexicon Learning for Wide-Coverage Semantic Parsing (C18-1)

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.

Copied to clipboard

Bo Chen, Bo An, Le Sun, Xianpei Han

Multimodality

Challenge:	Semantic parsers rely on accurate and high-coverage lexicons, but they often use annotated logical forms to learn the lexic.
Approach:	They propose a semi-supervised learning framework that makes use of large text corpora and lexical resources.
Outcome:	The proposed framework improves on two benchmarks: Webquestions and Free917.

Beyond Full Fine-tuning: Harnessing the Power of LoRA for Multi-Task Instruction Tuning (2024.lrec-main)

Copied to clipboard

Chunlei Xin, Yaojie Lu, Hongyu Lin, Shuheng Zhou, Huijia Zhu, Weiqiang Wang, Zhongyi Liu, Xianpei Han, Le Sun

Does the Correctness of Factual Knowledge Matter for Factual Knowledge-Enhanced Pre-trained Language Models? (2023.emnlp-main)

Challenge:	Low-Rank Adaptation (LoRA) is a parameter-efficient fine-tuning algorithm for large-scale language models.
Approach:	They conduct a systematic study of Low-Rank Adaptation (LoRA) on diverse tasks and rich resources with different learning capacities.
Outcome:	The proposed algorithm can achieve remarkable performance in high-resource and multi-task scenarios, even comparable to full fine-tuning.

Copied to clipboard

Boxi Cao, Qiaoyu Tang, Hongyu Lin, Xianpei Han, Le Sun

DeepSolution: Boosting Complex Engineering Solution Design via Tree-based Exploration and Bi-point Thinking (2025.acl-long)

Challenge:	Existing work neither proves that pre-trained models successfully learn the injected factual knowledge nor proves there is a causal relation between injected knowledge and downstream performance improvements.
Approach:	They propose a counterfactual-based analysis framework to explore the causal effects of factual knowledge injection on the performance of language models within pretrain-finetune paradigm.
Outcome:	The proposed framework shows that factual knowledge injection is successful but correctness of injected knowledge only has limited effect on the models’ downstream performance.

Copied to clipboard

Zhuoqun Li, Haiyang Yu, Xuanang Chen, Hongyu Lin, Yaojie Lu, Fei Huang, Xianpei Han, Yongbin Li, Le Sun

Challenge:	Existing studies in retrieval-augmented generation (RAG) do not sufficiently address the design of complex engineering solutions.
Approach:	They propose a retrieval-augmented generation system that leverages tree-based exploration and bi-point thinking mechanism to generate reliable solutions.
Outcome:	Experiments show that the proposed system achieves state-of-the-art (SOTA) performance on the SolutionBench, highlighting its potential to enhance the automation and reliability of complex engineering solution design in real-world applications.

Aligning Retrieval with Reader Needs: Reader-Centered Passage Selection for Open-Domain Question Answering (2025.coling-main)

Copied to clipboard

Chunlei Xin, Shuheng Zhou, Xuanang Chen, Yaojie Lu, Huijia Zhu, Weiqiang Wang, Zhongyi Liu, Xianpei Han, Le Sun

Model-Free Context-Aware Word Composition (C18-1)

Challenge:	Existing retrieval methods aim to gather relevant passages but fail to prioritize consistent and useful information for the reader.
Approach:	They propose a novel method which re-ranks passages based on the reader's prediction probability distribution and clusters passage according to the predicted answers.
Outcome:	The proposed method improves the quality of evidence passages under zero-shot scenarios.

Copied to clipboard

Bo An, Xianpei Han, Le Sun

When Models Outthink Their Safety: Unveiling and Mitigating Self-Jailbreak in Large Reasoning Models (2026.findings-acl)

Challenge:	Word composition models do not take ambiguity of words and context into consideration for learning representations, and thus suffer from the inaccurate representation of semantics.
Approach:	They propose a model-free context-aware word composition model which takes the latent semantic information as global context for learning representations.
Outcome:	The proposed model improves on existing composition models at different granularities and shows that it can be used to learn semantics.

Copied to clipboard

Yingzhi Mao, Chunkang Zhang, Junxiang Wang, Xinyan Guan, Boxi Cao, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun

Pre-training to Match for Unified Low-shot Relation Extraction (2022.acl-long)

Challenge:	Existing methods often apply coarse-grained constraints over entire reasoning trajectories . Existing approaches often apply unsafe constraints, causing unsafe outputs .
Approach:	They propose a trajectory-level training framework that mitigates Self-Jailbreak . they propose 'chain-of-guardrail' to mitigate self-jailbreak by targeting step-level interventions .
Outcome:	The proposed framework mitigates Self-Jailbreak by targeting step-level interventions while maintaining reasoning ability.

Copied to clipboard

Fangchao Liu, Hongyu Lin, Xianpei Han, Boxi Cao, Le Sun

BERT-QE: Contextualized Query Expansion for Document Re-ranking (2020.findings-emnlp)

Challenge:	Low-shot relation extraction (RE) aims to recognize novel relations with very few or even no samples.
Approach:	They propose a method that leverages triplet paraphrase to pre-train zero-shot label matching ability and uses meta-learning paradigm to learn few-shot instance summarizing ability.
Outcome:	The proposed method outperforms strong baselines and achieves the best performance on few-shot RE leaderboard.

Copied to clipboard

Zhi Zheng, Kai Hui, Ben He, Xianpei Han, Le Sun, Andrew Yates

CoCoNUTS: Concentrating on Content while Neglecting Uninformative Textual Styles for AI-Generated Peer Review Detection (2026.acl-long)

Challenge:	Existing methods to expand query use pseudo relevance feedback (PRF) but they are under-equipped to evaluate the relevance of information pieces used for expansion.
Approach:	They propose a query expansion model that leverages the BERT model to select relevant document chunks for expansion.
Outcome:	The proposed model significantly outperforms existing models on the TREC Robust04 and GOV2 test collections.

Copied to clipboard

Yihan Chen, Jiawei Chen, Guozhao Mo, Xuanang Chen, Ben He, Xianpei Han, Le Sun

Challenge:	Existing detectors rely on stylistic cues to distinguish between surface-level language refinement and genuine content generation.
Approach:	They propose a content-based detection paradigm to detect substantive AI-generation . they propose 'CoCoDet' detector that can detect surface-level language refinement .
Outcome:	The proposed detector achieves a macro F1 score of 98.24% on permissible machine-polished reviews and maintains 3.89% false positive rate on real-world reviews.

The Linguistic Connectivities Within Large Language Models (2025.findings-acl)

Copied to clipboard

Dan Wang, Boxi Cao, Ning Bian, Xuanang Chen, Yaojie Lu, Hongyu Lin, Jia Zheng, Le Sun, Shanshan Jiang, Bin Dong, Xianpei Han

READoc: A Unified Benchmark for Realistic Document Structured Extraction (2025.findings-acl)

Challenge:	Recent studies have discovered notable disparities in their performance across different languages.
Approach:	They conduct a systematic investigation into the behaviors of large language models across 27 different languages on 3 different scenarios and reveals a Linguistic Map correlates with the richness of available resources and linguistic family relations.
Outcome:	The proposed model demonstrates that there are significant disparities in performance across languages across 27 different languages on 3 different scenarios.

Copied to clipboard

Zichao Li, Aizier Abulaiti, Yaojie Lu, Xuanang Chen, Jia Zheng, Hongyu Lin, Xianpei Han, Shanshan Jiang, Bin Dong, Le Sun

Element Intervention for Open Relation Extraction (2021.acl-long)

Challenge:	Document Structured Extraction (DSE) is a field of document structure analysis that aims to extract structured content from raw documents.
Approach:	They propose a benchmark to evaluate document structured extraction systems by converting unstructured PDFs into semantically rich Markdown.
Outcome:	The proposed benchmark is based on 3,576 diverse and real-world documents from arXiv, GitHub, and Zenodo.

Copied to clipboard

Fangchao Liu, Lingyong Yan, Hongyu Lin, Xianpei Han, Le Sun

Text2Event: Controllable Sequence-to-Structure Generation for End-to-end Event Extraction (2021.acl-long)

Challenge:	Current OpenRE models are often trained on the datasets generated from distant supervision, which often results in instability and makes the model easily collapsed.
Approach:	They propose to use a causal model to identify relation instances referring to the same relation . they propose to perform Element Interventions on context and entities respectively .
Outcome:	The proposed method outperforms existing methods and is robust across datasets.

Copied to clipboard

Yaojie Lu, Hongyu Lin, Jin Xu, Xianpei Han, Jialong Tang, Annan Li, Le Sun, Meng Liao, Shaoyi Chen

XMC-Agent : Dynamic Navigation over Scalable Hierarchical Index for Incremental Extreme Multi-label Classification (2024.findings-acl)

Challenge:	Existing methods to extract event records from text decompose complex structure prediction task into multiple subtasks.
Approach:	They propose a sequence-to-structure generation paradigm that can extract events from text . they propose unified event extraction, constrained decoding algorithm and curriculum learning algorithm .
Outcome:	The proposed method can achieve competitive performance using record-level annotations in both supervised learning and transfer learning settings.

Copied to clipboard

Yanjiang Liu, Tianyun Zhong, Yaojie Lu, Hongyu Lin, Ben He, Shuheng Zhou, Huijia Zhu, Weiqiang Wang, Zhongyi Liu, Xianpei Han, Le Sun

Syntactic and Semantic-driven Learning for Open Information Extraction (2020.findings-emnlp)

Challenge:	Existing methods for XMC struggle with the growing set of labels due to their static label assumptions, and embedding-based methods struggle with complex mapping relationships due to late interaction paradigm.
Approach:	They propose a large language model (LLM) powered agent framework for extreme multi-label classification, XMC-Agent, which can effectively learn, manage and predict the extremely large and dynamically increasing set of labels.
Outcome:	The proposed framework can learn, manage and predict the extremely large and dynamically growing set of labels and achieves state-of-the-art performance on three standard datasets.

Copied to clipboard

Jialong Tang, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun, Xinyan Xiao, Hua Wu

Meta-Cognitive Analysis: Evaluating Declarative and Procedural Knowledge in Datasets and Large Language Models (2024.lrec-main)

Challenge:	Experimental results show that our approach significantly outperforms the supervised counterparts, and can even achieve competitive performance to supervised state-of-the-art (SoA) model.
Approach:	They propose a syntactic and semantic-driven learning approach that can learn open IE models without human-labelled data by leveraging syntakic and semantic knowledge as noisier, higher-level supervision.
Outcome:	The proposed approach outperforms supervised counterparts and can achieve competitive performance to supervised state-of-the-art models.

Copied to clipboard

Zhuoqun Li, Hongyu Lin, Yaojie Lu, Hao Xiang, Xianpei Han, Le Sun

Debiasing In-Context Learning by Instructing LLMs How to Follow Demonstrations (2024.findings-acl)

Challenge:	Recent advances in large language models have push NLP into a new era, moving away from traditional task-specific pre-train finetuning paradigm.
Approach:	They provide a comprehensive analysis of declarative and procedural knowledge for large language models and evaluate their effectiveness.
Outcome:	The proposed model can perform better with both kinds of knowledge, but at different speeds.

Copied to clipboard

Lvxue Li, Jiaqi Chen, Xinyu Lu, Yaojie Lu, Hongyu Lin, Shuheng Zhou, Huijia Zhu, Weiqiang Wang, Zhongyi Liu, Xianpei Han, Le Sun

Match, Compare, or Select? An Investigation of Large Language Models for Entity Matching (2025.coling-main)

Challenge:	In-context learning (ICL) has gained considerable attention due to its data efficiency and task adaptability.
Approach:	They propose to de-biase demonstration bias in in-context learning by focusing on semantic ambiguity induced by demonstrations and reducing the semantic hazard.
Outcome:	The proposed methods significantly improve performance on six datasets.

Copied to clipboard

Tianshu Wang, Xiaoyang Chen, Hongyu Lin, Xuanang Chen, Xianpei Han, Le Sun, Hao Wang, Zhenyu Zeng

Self-Steering Optimization: Autonomous Preference Optimization for Large Language Models (2025.findings-acl)

Challenge:	Entity matching (EM) is a critical step in entity resolution (ER).
Approach:	They propose a method that incorporates record interactions from different perspectives.
Outcome:	The proposed framework improves on 8 ER datasets and 10 LLMs and achieves higher efficiency and effectiveness.

Copied to clipboard

Hao Xiang, Bowen Yu, Hongyu Lin, Keming Lu, Yaojie Lu, Xianpei Han, Ben He, Le Sun, Jingren Zhou, Junyang Lin

ShortGPT: Layers in Large Language Models are More Redundant Than You Expect (2025.findings-acl)

Challenge:	Prior research focused on developing data generation methods, while insufficient attention has been paid to quality control mechanisms and often produces inaccurate and unhelpful data.
Approach:	They propose an algorithm that automatically generates high-quality preference data, eliminating manual annotation requirements.
Outcome:	The proposed algorithm outperforms baselines in human preference alignment and reward optimization.

Copied to clipboard

Xin Men, Mingyu Xu, Qingyu Zhang, Qianhao Yuan, Bingning Wang, Hongyu Lin, Yaojie Lu, Xianpei Han, Weipeng Chen

Transferable Post-training via Inverse Value Learning (2025.naacl-long)

Challenge:	Recent studies have identified significant redundancy in large language models . quantization and pruning are two methods that reduce computational resources .
Approach:	They propose simple pruning methods that prune redundant layers based on their BI scores.
Outcome:	The proposed pruning methods demonstrate superior performance over previous pruning methods.

Copied to clipboard

Xinyu Lu, Xueru Wen, Yaojie Lu, Bowen Yu, Hongyu Lin, Haiyang Yu, Le Sun, Xianpei Han, Yongbin Li

Rule or Story, Which is a Better Commonsense Expression for Talking with Large Language Models? (2024.acl-long)

Challenge:	Existing algorithms for post-training large datasets are requiring a large computational effort.
Approach:	They propose to model the changes at logits level during post-training using a separate neural network . they demonstrate that the value network can be seamlessly integrated with another pre-trained model .
Outcome:	The proposed model can be integrated with another pre-trained model during inference, enabling similar capability enhancements.

Copied to clipboard

Ning Bian, Xianpei Han, Hongyu Lin, Yaojie Lu, Ben He, Le Sun

On the Editability of Delta Parameters in Post-Trained Models (2026.findings-acl)

Challenge:	Experimental results show that stories outperform rules as the expression for retrieving commonsense from LLMs, exhibiting higher generation confidence and commonsensense accuracy.
Approach:	They investigate the commonsense ability of large language models expressed through stories and rules to retrieve commonsensing knowledge from LLMs.
Outcome:	The stories outperform rules as commonsense expressions on 28 commonsensense QA datasets, exhibiting higher generation confidence and commonsence accuracy.

Copied to clipboard

Qiaoyu Tang, Le Yu, Bowen Yu, Hongyu Lin, Keming Lu, Yaojie Lu, Xianpei Han, Le Sun

AutoAlign: Get Your LLM Aligned with Minimal Annotations (2025.acl-demo)

Challenge:	Several studies have explored delta parameter properties via pruning, quantization, low-rank approximation, and extrapolation, but what properties of delta parameters are essential for maintaining performance?
Approach:	They propose to examine delta parameter properties along magnitude and sign . they propose to use a loss-based local surrogate analysis to examine editing effects .
Outcome:	The proposed analysis shows that delta parameters can be edited while maintaining performance.

Copied to clipboard

Xinyu Lu, Dong Xu, Chunkang Zhang, Xinyan Guan, Junxiang Wang, Qingyu Zhang, Pengbo Wang, Yingzhi Mao, Hao Xiang, Xueru Wen, Zichao Li, Yaojie Lu, Hongyu Lin, Le Sun, Xianpei Han

EUSP: An Easy-to-Use Semantic Parsing PlatForm (D19-3)

Challenge:	Automated Alignment (ALM) is a set of algorithms designed to align Large Language Models (LLMs) with human intentions and values while minimizing manual intervention.
Approach:	They propose an open-source toolkit that integrates mainstream automated algorithms through a consistent interface and an accessible workflow supporting one-click execution for prompt synthesis and automatic alignment signal construction.
Outcome:	The proposed framework enables easy reproduction of existing results through extensive benchmarks and facilitates the development of novel approaches via modular components.

Copied to clipboard

Bo An, Chen Bo, Xianpei Han, Le Sun

Expanding the Boundaries of Vision Prior Knowledge in Multi-modal Large Language Models (2026.eacl-long)

Challenge:	Semantic parsing aims to map natural language utterances into structured meaning representations.
Approach:	They propose a modular platform that allows developers to build semantic parser from scratch.
Outcome:	The proposed platform achieves competitive performance on semantic parsing task and improves performance of a business search engine.

Copied to clipboard

Qiao Liang, Yanjiang Liu, Weixiang Zhou, Ben He, Yaojie Lu, Hongyu Lin, Jia Zheng, Xianpei Han, Le Sun, Yingfei Sun

Multimodality

Challenge:	Existing research treats MLLMs as unified systems optimized through end-to-end training, but the impact of vision encoder’s prior knowledge is seldom investigated.
Approach:	They propose a metric to quantify the effect of prior knowledge on MLLM performance by integrating prior knowledge at the vision encoder level into a training framework.
Outcome:	The proposed training framework incorporates prior knowledge at the vision encoder level, and significantly boosts visual understanding capabilities of MLLMs.

Knowing When to Quit: Diagnosing and Training LLMs to Abort Futile Reasoning (2026.findings-acl)

Copied to clipboard

Xinyan Guan, Jiali Zeng, Chunlei Xin, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun, Fandong Meng

Open Grounded Planning: Challenges and Benchmark Construction (2024.acl-long)

Challenge:	Large language models generate costly yet semantically void reasoning on beyond-capability tasks . the dominant failure mode is specious reasoning, superficially valid outputs with subtle hallucinations .
Approach:	They propose a capability-aligned reinforcement learning approach that aligns model behavior with capability boundaries.
Outcome:	The proposed model reduces futile reasoning while maintaining performance across tasks.

Copied to clipboard

Shiguang Guo, Ziliang Deng, Hongyu Lin, Yaojie Lu, Xianpei Han, Le Sun

Challenge:	Existing work on LLM-based planning uses language generation to produce free-style plans . however, these plans are not grounded in an executable set of actions .
Approach:	They propose a new task for open grounded planning that asks the model to generate an executable plan based on a variable action set.
Outcome:	The proposed task is open grounded planning, which is based on a set of variables.

PPTAgent: Generating and Evaluating Presentations Beyond Text-to-Slides (2025.emnlp-main)

Copied to clipboard

Hao Zheng, Xinyan Guan, Hao Kong, Wenkai Zhang, Jia Zheng, Weixiang Zhou, Hongyu Lin, Yaojie Lu, Xianpei Han, Le Sun

Multimodality

Challenge:	Existing methods for generating presentations from documents focus on improving and evaluating content quality in isolation, overlooking visual appeal and structural coherence.
Approach:	They propose an edit-based presentation generation system that analyzes and iterates on slides to create new slides.
Outcome:	The proposed presentation generation tool outperforms existing methods in three dimensions . it analyzes slides, iterates and generates edit actions based on selected slides .

Learning to Bootstrap for Entity Set Expansion (D19-1)

Copied to clipboard

Lingyong Yan, Xianpei Han, Le Sun, Ben He

Seg2Act: Global Context-aware Action Generation for Document Logical Structuring (2024.emnlp-main)

Challenge:	Existing bootstrapping methods for Entity Set Expansion suffer from two problems: 1) delayed feedback and sparse supervision.
Approach:	They propose a method that estimates delayed feedback and adaptively scores entities given sparse supervision signals.
Outcome:	The proposed method can estimate delayed feedback for pattern evaluation and adaptively score entities given sparse supervision signals.

Copied to clipboard

Zichao Li, Shaojie He, Meng Liao, Xuanang Chen, Yaojie Lu, Hongyu Lin, Yanxiong Lu, Xianpei Han, Le Sun

Challenge:	Document logical structuring is crucial for document intelligence due to the complexity of text segment dependencies in the document.
Approach:	They propose an end-to-end, generation-based method for document logical structuring that generates the action sequence via a global context-aware generative model and updates its global context and current logical structure based on the generated actions.
Outcome:	Experiments on ChCatExt and HierDoc datasets show that Seg2Act performs better than previous methods in both supervised and transfer learning settings.

Can Prompt Probe Pretrained Language Models? Understanding the Invisible Risks from a Causal View (2022.acl-long)

Copied to clipboard

Boxi Cao, Hongyu Lin, Xianpei Han, Fangchao Liu, Le Sun

Sequence-to-Nuggets: Nested Entity Mention Detection via Anchor-Region Networks (P19-1)

Challenge:	Recent studies have found prompt-based probing evaluations inaccurate, inconsistent and unreliable.
Approach:	They propose to conduct debiasing via causal intervention to uncover biases in probing evaluations . authors argue that prompt-based probing is inaccurate, inconsistent and unreliable .
Outcome:	This paper examines the effectiveness of prompt-based probing in pretrained language models . it highlights critical biases which could induce biased results and conclusions . authors suggest rethinking criteria for evaluating better pretrained models based on such evaluations .

Copied to clipboard

Fine-grained Entity Typing via Label Reasoning (2021.emnlp-main)

Challenge:	Named entity recognition (NER) approaches restrict each word belonging to at most one entity mention.
Approach:	They propose to model and leverage the head-driven phrase structures of entity mentions to solve this problem.
Outcome:	The proposed architecture achieves state-of-the-art on three standard nested entity mention detection benchmarks.

Copied to clipboard

Qing Liu, Hongyu Lin, Xinyan Xiao, Xianpei Han, Le Sun, Hua Wu

From Discourse to Narrative: Knowledge Projection for Event Relation Extraction (2021.acl-long)

Challenge:	Existing approaches to fine-grained entity typing are based on independent classification paradigms, which make them difficult to recognize inter-dependent, long-tailed and fine-granular entities.
Approach:	They propose a label reasoning network that exploits label dependencies knowledge entailed in the data.
Outcome:	The proposed network can model, learn and reason complex labels in a sequence-to-set, end-to end manner.

Copied to clipboard

Jialong Tang, Hongyu Lin, Meng Liao, Yaojie Lu, Xianpei Han, Le Sun, Weijian Xie, Jin Xu

WebDP: Understanding Discourse Structures in Semi-Structured Web Documents (2023.findings-acl)

Challenge:	Existing event-centric knowledge graphs rely on explicit connectives to extract relations between events.
Approach:	They propose a knowledge projection paradigm for event relation extraction using commonalities between events.
Outcome:	The proposed method achieves state-of-the-art performance and extrinsic results verify the extracted event relations.

Copied to clipboard

Peilin Liu, Hongyu Lin, Meng Liao, Hao Xiang, Xianpei Han, Le Sun

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

Challenge:	Web documents are one of the most primary and biggest data resources in current era, and understanding their discourse structure will benefit various downstream document processing applications.
Approach:	They propose a web document discourse structure representation schema by extending classical discourse theories and adding special features to well represent discourse characteristics of web documents.
Outcome:	The proposed task is feasible but challenging for current models.

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.

Chain-of-Rewrite: Aligning Question and Documents for Open-Domain Question Answering (2024.findings-emnlp)

Copied to clipboard

Chunlei Xin, Yaojie Lu, Hongyu Lin, Shuheng Zhou, Huijia Zhu, Weiqiang Wang, Zhongyi Liu, Xianpei Han, Le Sun

Natural Language Interfaces

Challenge:	Existing approaches to answer open-domain question have encountered term mismatch and limited interaction between IR systems and large language models.
Approach:	They propose a method which leverages the guidance and feedback gained from the analysis to provide faithful and consistent extensions for effective question answering.
Outcome:	Experiments on four open-domain question answering datasets show the proposed method performs well under zero-shot settings.

Contrastive Distant Supervision for Debiased and Denoised Machine Reading Comprehension (2023.findings-emnlp)

Copied to clipboard

Ning Bian, Hongyu Lin, Xianpei Han, Ben He, Le Sun

Accurate Text-Enhanced Knowledge Graph Representation Learning (N18-1)

Challenge:	Distant supervision (DS) is a promising learning approach for machine reading comprehension (MRC) however, the annotated dataset will inevitably lead to mislabeled instances, resulting in answer bias and context noise problems.
Approach:	They propose an algorithm that can learn to distinguish confusing and noisy instances via confidence-aware contrastive learning.
Outcome:	The proposed algorithm can learn to distinguish confusing and noisy instances via confidence-aware contrastive learning.

Copied to clipboard

Bo An, Bo Chen, Xianpei Han, Le Sun

Global Bootstrapping Neural Network for Entity Set Expansion (2020.findings-emnlp)

Challenge:	Existing representation learning methods for knowledge graph representation do not consider the ambiguity of relations and entities.
Approach:	They propose a text-enhanced knowledge graph representation learning method which exploits the entity descriptions and triple-specific relation mention to enhance representations.
Outcome:	The proposed method outperforms existing representation learning models on link prediction and triple classification tasks and significantly outperformed existing models.

Copied to clipboard

Lingyong Yan, Xianpei Han, Ben He, Le Sun

Cost-sensitive Regularization for Label Confusion-aware Event Detection (P19-1)

Challenge:	Recent studies have shown that end-to-end bootstrapping methods only leverage local semantics rather than global semantics.
Approach:	They propose a global-sighted encoder to capture and encode local and global semantics into entity embedding and an attention-guided decoder to sequentially expand new entities based on these embeddables.
Outcome:	The proposed network achieves state-of-the-art on two bootstrapping datasets.

Copied to clipboard

Gazetteer-Enhanced Attentive Neural Networks for Named Entity Recognition (D19-1)

Challenge:	Recent advances in event detection focus on wordwise classification with one NIL class for tokens do not trigger any event.
Approach:	They propose a cost-sensitive regularization method which penalizes more on mislabeling . they propose two estimators which can effectively measure such label confusion based on instance-level statistics .
Outcome:	The proposed method can improve the performance of different models in English and Chinese event detection.

Copied to clipboard

Hongyu Lin, Yaojie Lu, Xianpei Han, Le Sun, Bin Dong, Shanshan Jiang

Progressive Adversarial Learning for Bootstrapping: A Case Study on Entity Set Expansion (2021.emnlp-main)

Challenge:	Named entity recognition (NER) is a fundamental NLP task.
Approach:	They propose a gazetteer-based attentive neural network which can enhance region-based NER . they first model the mention-context association and then an auxiliary gazetteers .
Outcome:	The proposed approach can achieve state-of-the-art on ACE2005 named entity recognition benchmark.

Copied to clipboard

Lingyong Yan, Xianpei Han, Le Sun

Robustness in NLP

Challenge:	Existing methods for entity set expansion define the expansion boundary using seed-based distance metrics, which are hard to adjust due to the extremely sparse supervision.
Approach:	They propose a new learning method for bootstrapping which jointly models the bootstraping process and boundary learning process in a GAN framework.
Outcome:	The proposed method achieves the new state-of-the-art performance for entity set expansion.

Nugget Proposal Networks for Chinese Event Detection (P18-1)

Copied to clipboard

CRUXEVAL-X: A Benchmark for Multilingual Code Reasoning, Understanding and Execution (2025.acl-long)

Challenge:	Nugget Proposal Networks (NPNs) can solve word-trigger mismatch problem . word-wise event detection models suffer from word-tree mismatch because of multiple triggers .
Approach:	They propose a novel way to detect event triggers in a character-wise paradigm . they propose entire trigger nuggets centered at each character regardless of word boundaries .
Outcome:	The proposed model outperforms the state-of-the-art methods on two datasets.

Copied to clipboard

Ruiyang Xu, Jialun Cao, Yaojie Lu, Ming Wen, Hongyu Lin, Xianpei Han, Ben He, Shing-Chi Cheung, Le Sun

Programming Languages in NLP

Challenge:	Existing code benchmarks focus on code generation, while those for code reasoning are insufficient.
Approach:	They propose a multi-lingual code reasoning benchmark that contains 19 programming languages and at least 600 subjects for each language.
Outcome:	The proposed model trains on Python and achieves 34.4% Pass@1 in other languages, revealing the cross-language generalization of LLMs.

StructEval: Deepen and Broaden Large Language Model Assessment via Structured Evaluation (2024.findings-acl)

Copied to clipboard

Boxi Cao, Mengjie Ren, Hongyu Lin, Xianpei Han, Feng Zhang, Junfeng Zhan, Le Sun

Sparse Latents Steer Retrieval-Augmented Generation (2025.acl-long)

Challenge:	Current evaluations for large language models use a single-item assessment paradigm . current evaluations struggle to discern whether a model possesses the required capabilities or merely memorizes/guesses the answers to specific questions.
Approach:	They propose a framework to evaluate large language models using atomic test objectives.
Outcome:	The proposed evaluation framework resists data contamination and reduces interference of potential biases, and sheds light on the design of future principled and trustworthy LLM evaluation protocols.

Copied to clipboard

Chunlei Xin, Shuheng Zhou, Huijia Zhu, Weiqiang Wang, Xuanang Chen, Xinyan Guan, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun

ECO v1: Towards Event-Centric Opinion Mining (2022.findings-acl)

Challenge:	In this study, we uncover interpretable latents that govern RAG behavior in large language models . Sparse Autoencoders are used to control large language model (LLM) behavior .
Approach:	They leverage Sparse Autoencoders within the LLaMA Scope to uncover latents that govern RAG behaviors.
Outcome:	The proposed model can be used to control large language models without architectural modifications.

Copied to clipboard

Ruoxi Xu, Hongyu Lin, Meng Liao, Xianpei Han, Jin Xu, Wei Tan, Yingfei Sun, Le Sun

Sentiment Analysis

Challenge:	Existing studies on event-centric opinion mining focus on entity-centric opinions . entity-centered opinions focus on sentimental polarity of events, while event-centered ones focus on content .
Approach:	They propose to perform event-centric opinion mining on event-argument structure and expression categorizing theory and benchmark it against a pioneer corpus.
Outcome:	The proposed task is feasible and challenging, and the results are beneficial for future studies.

Ambiguous Learning from Retrieval: Towards Zero-shot Semantic Parsing (2023.acl-long)

Copied to clipboard

Shan Wu, Chunlei Xin, Hongyu Lin, Xianpei Han, Cao Liu, Jiansong Chen, Fan Yang, Guanglu Wan, Le Sun

Improved Sparse Upcycling for Instruction Tuning (2025.coling-main)

Challenge:	Existing neural semantic parsers require a large amount of training data which is expensive and difficult to obtain.
Approach:	They propose a framework for a supervised retrieval system based on pretrained language models . they propose ambiguous supervision to improve the precision and coverage of the task .
Outcome:	The proposed approach outperforms state-of-the-art zero-shot parsing methods in ambiguous supervision.

Copied to clipboard

Wangyi Jiang, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun

SoFA: Shielded On-the-fly Alignment via Priority Rule Following (2024.findings-acl)

Challenge:	Existing methods for sparse upcycling lead to performance degradation in instruction tuning scenarios.
Approach:	They propose a representation-based approach to convert dense language models into sparsely activated ones by initializing router weights from language models.
Outcome:	The proposed architecture improves model capabilities and routing consistency across multiple benchmarks.

Copied to clipboard

Xinyu Lu, Bowen Yu, Yaojie Lu, Hongyu Lin, Haiyang Yu, Le Sun, Xianpei Han, Yongbin Li