Papers by Gu Xu

Beyond Demographics: Enhancing Cultural Value Survey Simulation with Multi-Stage Personality-Driven Cognitive Reasoning (2025.emnlp-main)

Challenge:	Prior work synthesizes tool-use LLM datasets by first generating a user query, then complex tool-using annotations like DFS.
Approach:	They propose an agentic framework that synthesizes user queries and generates valid tool-use chains . they propose a dataset with more complex tool use, lower cost, and almost 100% pass rate .
Outcome:	Experiments show that tools trained on ToolGrad outperform expensive baseline datasets and proprietary LLMs.

Copied to clipboard

Haijiang Liu, Qiyuan Li, Chao Gao, Yong Cao, Xiangyu Xu, Xun Wu, Daniel Hershcovich, Jinguang Gu

NeuronMerge: Merging Models via Functional Neuron Groups (2025.findings-acl)

Challenge:	Introducing MARK, a framework for cultural value survey simulation . based on type dynamics theory, it improves accuracy and interpretation of models .
Approach:	They propose a framework that integrates psychological theory into cultural value survey simulations.
Outcome:	The proposed framework outperforms baseline models on the World Values Survey by 10% accuracy and reduces divergence between model predictions and human preferences.

Copied to clipboard

Wangyun Gu, Qianghua Gao, Zhang Li-Xin, Xu Shen, Jieping Ye

Adapting Meta Knowledge Graph Information for Multi-Hop Reasoning over Few-Shot Relations (D19-1)

Challenge:	Recent studies show that task arithmetic improves performance by combining model parameters with output features.
Approach:	They propose a neuron-based task arithmetic merging method that improves model linearity . they group neurons by function and propose combining them with existing models .
Outcome:	The proposed method improves performance across tasks and scales.

Copied to clipboard

Xin Lv, Yuxian Gu, Xu Han, Lei Hou, Juanzi Li, Zhiyuan Liu

Non-Autoregressive Sequence Generation (2022.acl-tutorials)

Challenge:	Existing methods for multi-hop reasoning assume that every relation has enough triples for training . however, performance drops significantly on few-shot relations .
Approach:	They propose a meta-based multi-hop reasoning method that learns meta parameters from high-frequency relations that could quickly adapt to few-shot scenarios.
Outcome:	The proposed method outperforms state-of-the-art methods in few-shot scenarios on two public datasets from Freebase and NELL.

Copied to clipboard

Jiatao Gu, Xu Tan

TextLap: Customizing Language Models for Text-to-Layout Planning (2024.findings-emnlp)

Challenge:	Non-autoregressive sequence generation (NAR) models generate output sequences in parallel to speed up generation process.
Approach:	This tutorial provides a thorough introduction and review of non-autoregressive sequence generation . it aims to generate the entire or partial output sequences in parallel to speed up the generation process .
Outcome:	This tutorial provides a thorough introduction and review of non-autoregressive sequence generation . it aims to reduce the performance gap between state-of-the-art models due to lack of modeling power .

Copied to clipboard

Jian Chen, Ruiyi Zhang, Yufan Zhou, Jennifer Healey, Jiuxiang Gu, Zhiqiang Xu, Changyou Chen

Constraining Sequential Model Editing with Editing Anchor Compression (2025.findings-naacl)

Challenge:	Creating 2D graphical layouts from text alone is challenging in traditional settings.
Approach:	They propose to customize LLMs to allow users to generate professional looking layouts by simply inputting text instructions.
Outcome:	The proposed method outperforms existing benchmarks for document generation and graphical design benchmarks.

Copied to clipboard

Hao-Xiang Xu, Jun-Yu Ma, Zhen-Hua Ling, Ningyu Zhang, Jia-Chen Gu

MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsing (2026.acl-industry)

Challenge:	Large language models (LLMs) exhibit hallucinations due to incorrect or outdated knowledge embedded in their parameters.
Approach:	They propose a framework to constrain the deviation of the parameter matrix during sequential editing by selecting editing anchors that are important in encoding new relations without deviating too much from the original matrix.
Outcome:	The proposed framework minimizes deviations caused by model editing while retaining over 70% of the general abilities.

Copied to clipboard

Junbo Niu, Zheng Liu, Zhuangcheng Gu, Bin Wang, Linke Ouyang, Zhiyuan Zhao, Tao Chu, Tianyao He, Fan Wu, Qintong Zhang, Zhenjiang Jin, Guang Liang, Rui Zhang, Wenzheng Zhang, Yuan Qu, Zhifei Ren, Yuefeng Sun, Zirui Tang, Boyu Niu, Yuanhong Zheng, Dongsheng Ma, Ziyang Miao, Hejun Dong, Siyi Qian, Junyuan Zhang, Fangdong Wang, Jingzhou Chen, Xiaomeng Zhao, Liqun Wei, Wei Li, Shasha Wang, RuiLiang Xu, Yuanyuan Cao, Lu Chen, Qianqian Wu, Huaiyu Gu, Lindong Lu, Dechen Lin, null Shenguanlin, Xuanhe Zhou, Linfeng Zhang, Yuhang Zang, Xiaoyi Dong, Jiaqi Wang, Bo Zhang, Lei Bai, Pei Chu, Weijia Li, Jiang Wu, Lijun Wu, Zhenxiang Li, Guangyu Wang, Zhongying Tu, Chao Xu, Kai Chen, Bowen Zhou, Dahua Lin, Wentao Zhang, Conghui He

Green & Sustainable NLP

Challenge:	Document images are characterized by higher resolutions, denser content, and more complex structural layouts.
Approach:	They propose a 1.2B-parameter document parsing vision-language model that decouples layout analysis from local content recognition.
Outcome:	The proposed model surpasses general-purpose and domain-specific models on multiple benchmarks while maintaining significantly lower computational overhead.

Benchmarking Open-ended Audio Dialogue Understanding for Large Audio-Language Models (2025.acl-long)

Copied to clipboard

Kuofeng Gao, Shu-Tao Xia, Ke Xu, Philip Torr, Jindong Gu

Dialogue Systems & Conversational Agents

Challenge:	Large Audio-Language Models (LALMs) have recently unlocked audio dialogue capabilities, enabling direct spoken exchanges with humans.
Approach:	They propose to evaluate LALMs' open-ended audio dialogue ability in 3 general scenarios, 12 skills, 9 multilingual languages, and 4 categories of ambiguity handling.
Outcome:	The proposed benchmark assesses the open-ended audio dialogue ability for LALMs in 3 general scenarios, 12 skills, 9 multilingual languages, and 4 categories of ambiguity handling.

AlignBench: Benchmarking Chinese Alignment of Large Language Models (2024.acl-long)

Copied to clipboard

Xiao Liu, Xuanyu Lei, Shengyuan Wang, Yue Huang, Andrew Feng, Bosi Wen, Jiale Cheng, Pei Ke, Yifan Xu, Weng Lam Tam, Xiaohan Zhang, Lichao Sun, Xiaotao Gu, Hongning Wang, Jing Zhang, Minlie Huang, Yuxiao Dong, Jie Tang

Robustness of Learning from Task Instructions (2023.findings-acl)

Challenge:	Effective evaluation of alignment for emerging Chinese LLMs is still significantly lacking, calling for real-scenario grounded, open-ended, challenging and automatic evaluations tailored for alignment.
Approach:	They propose a multi-dimensional benchmark for evaluating LLMs’ alignment in Chinese with 8 main categories, 683 real-scenario rooted queries and corresponding human verified references.
Outcome:	The benchmark uses a human-in-the-loop data curation pipeline, 683 real-scenario rooted queries and human verified references.

Copied to clipboard

Jiasheng Gu, Hongyu Zhao, Hanzi Xu, Liangyu Nie, Hongyuan Mei, Wenpeng Yin

Robustness in NLP

Challenge:	traditional supervised learning mostly works on individual tasks and requires training on a large set of task-specific examples.
Approach:	a new study investigates the system robustness when instructions are manipulated and paraphrased . task instructions give the model the definition of the task and allow it to output the appropriate answer .
Outcome:	a new study shows that supervised learning is robust when instructions are manipulated, paraphrased or iii from different levels of conciseness.

Bit-by-Bit: Progressive QAT Strategy with Outlier Channel Splitting for Stable Low-Bit LLMs (2026.acl-long)

Copied to clipboard

Binxing Xu, Hao Gu, Lujun Li, Hao Wang, Bei Liu, Jiacheng Liu, Qiyuan Zhu, Xintong Yang, Chao Li, Sirui Han, Yike Guo

Green & Sustainable NLP

Challenge:	Existing approaches to training LLMs at ultra-low precisions suffer from convergence instability and substantial training costs.
Approach:	They propose a progressive QAT framework with outlier channel splitting to address these issues . they use nested structure of integer quantization grids to enable a "train once, deploy any precision" paradigm .
Outcome:	The proposed framework outperforms baselines on both Llama2/3 and W2A16, with an 11 speedup over BF16.

PersLEARN: Research Training through the Lens of Perspective Cultivation (2023.acl-demo)

Copied to clipboard

Yu-Zhe Shi, Shiqian Li, Xinyi Niu, Qiao Xu, Jiawen Liu, Yifan Xu, Shiyu Gu, Bingru He, Xinyang Li, Xinyu Zhao, Zijian Zhao, Yidong Lyu, Zhen Li, Sijia Liu, Lin Qiu, Jinhao Ji, Lecheng Ruan, Yuxi Ma, Wenjuan Han, Yixin Zhu

Challenge:	PersLEARN is a tool designed to facilitate the cultivation of scientific perspectives . junior researchers struggle to identify the perspectives reflected in the literature and struggle to develop their own viewpoints.
Approach:	They propose a tool to facilitate the cultivation of scientific perspectives by interacting with a prompt-based model and allowing students to develop their own perspectives explicitly.
Outcome:	The proposed tool outperforms baseline approaches across multiple domains of literature from different perspectives.

RealHiTBench: A Comprehensive Realistic Hierarchical Table Benchmark for Evaluating LLM-Based Table Analysis (2025.findings-acl)

Copied to clipboard

Pengzuo Wu, Yuhang Yang, Guangcheng Zhu, Chao Ye, Hong Gu, Xu Lu, Ruixuan Xiao, Bowen Bao, Yijing He, Liangyu Zha, Wentao Ye, Junbo Zhao, Haobo Wang

Investigating and Mitigating the Multimodal Hallucination Snowballing in Large Vision-Language Models (2024.acl-long)

Challenge:	Existing benchmarks for large language models focus on simple, flat table structures.
Approach:	They propose a benchmark to evaluate the performance of both Large Language Models and Multimodal LLMs across a variety of input formats for complex tabular data, including LaTeX, HTML, and PNG.
Outcome:	The proposed benchmark evaluates the performance of LLMs and Multimodal LLM models across a variety of input formats for complex tabular data, including LaTeX, HTML, and PNG.

Copied to clipboard

Weihong Zhong, Xiaocheng Feng, Liang Zhao, Qiming Li, Lei Huang, Yuxuan Gu, Weitao Ma, Yuan Xu, Bing Qin

POP-CEE: Position-oriented Prompt-tuning Model for Causal Emotion Entailment (2024.findings-acl)

Challenge:	Large Vision-Language Models (LVLMs) suffer from multimodal hallucinations . however, the generated hallucines could influence the models’ subsequent generation .
Approach:	They propose a framework to evaluate LVLMs' behaviors when encountering generated hallucinations and a method to revise the output distribution of LVLs with the one derived from the residual visual input.
Outcome:	The proposed framework reduces the performance of open-source LVLMs by 31%, indicating that they are prone to accept the generated hallucinations and make false claims that they would not have supported without distractions.

Copied to clipboard

Zhihan Zhou, Xue Gu, Yujie Zhao, Hao Xu

Mitigating Spurious Correlations in Text Classification Using Latent Space Geometry (2026.acl-long)

Challenge:	Existing methods for emotion analysis in conversations ignore the specific semantic associations between emotions and cause utterances.
Approach:	They propose a position-oriented prompt-tuning model to solve the CEE task in an end-to-end manner.
Outcome:	The proposed model achieves state-of-the-art performance on a benchmark dataset.

Copied to clipboard

Jiasen Gao, Xiaoliang Chen, Duoqian Miao, Xu Gu, Xianyong Li, Yajun Du

Text Classification

Challenge:	Existing models rely on predictive shortcuts that hold in training data but break under distribution shifts, leading to large performance drops for minority groups.
Approach:	They propose a framework that transforms abstract biases into interpretable geometric anchors without auxiliary classifiers by manipulating latent space geometry.
Outcome:	The proposed framework outperforms state-of-the-art baselines and improves worst-group accuracy by over 20% on the CivilComments dataset.

PPT: Pre-trained Prompt Tuning for Few-shot Learning (2022.acl-long)

Copied to clipboard

Yuxian Gu, Xu Han, Zhiyuan Liu, Minlie Huang

Anchored Cyclic Generation: A Novel Paradigm for Long-Sequence Symbolic Music Generation (2026.findings-acl)

Challenge:	Prompt tuning for pre-trained language models has shown remarkable performance . however, prompt tuning is still not fully explored .
Approach:	They propose to pre-train prompts by adding soft prompts into the pre-training stage to obtain a better initialization.
Outcome:	The proposed framework outperforms full-model tuning under full-data and few-shot learning settings.

Copied to clipboard

Boyu Cao, Lekai Qian, Dehan Li, Haoyu Gu, Mingda Xu, Qi Liu

From Long Videos to Engaging Clips: A Human-Inspired Video Editing Framework with Multimodal Narrative Understanding (2025.emnlp-industry)

Challenge:	Autoregressive sequence modeling has been successful in many domains, but maintaining long-term coherence and structural integrity remains a challenge.
Approach:	They propose an ACG paradigm that relies on anchor features from previously generated musical content to guide subsequent generation during the autoregressive process.
Outcome:	The proposed framework outperforms existing methods in symbolic music generation tasks.

Copied to clipboard

Xiangfeng Wang, Xiao Li, Yadong Wei, null Songxueyu, Yang Song, null Xiaxiaoqiang, Fangrui Zeng, Zaiyi Chen, null Liuliu, Gu Xu, Tong Xu

Mitigating Judgment Preference Bias in Large Language Models through Group-Based Polling (2026.findings-acl)

Challenge:	Existing methods for video editing rely on textual cues from ASR transcripts and segment selection, often neglecting rich visual context.
Approach:	They propose a human-inspired automatic video editing framework that leverages multimodal narrative understanding to address these limitations.
Outcome:	The proposed framework outperforms existing baselines across general and advertisement-oriented editing tasks.

Copied to clipboard

Shuliang Liu, Zhipeng Xu, Zhenghao Liu, Yukun Yan, Minghe Yu, Yu Gu, Chong Chen, Huiyuan Xie, Ge Yu

StructFact: Reasoning Factual Knowledge from Structured Data with Large Language Models (2025.findings-acl)

Challenge:	Large Language Models (LLMs) are used as automatic evaluators to provide accurate and reliable assessments.
Approach:	They propose a framework that integrates LLM-based judgment models into a multi-agent system and simulates the interactive client-server polling mechanism.
Outcome:	The proposed framework outperforms supervised models trained on annotated judgment data while requiring no human-labeled annotations.

Copied to clipboard

Sirui Huang, Yanggan Gu, Zhonghao Li, Xuming Hu, Li Qing, Guandong Xu

Leveraging Large Language Models for NLG Evaluation: Advances and Challenges (2024.emnlp-main)

Challenge:	Large language models (LLMs) have made significant strides in natural language processing by leveraging their ability to comprehend and reason with factual knowledge.
Approach:	They propose a benchmark to evaluate the ability of large language models to reason with structured data for knowledge-intensive tasks.
Outcome:	Extensive tests on 10 common LLMs show that they struggle with heterogeneity of structured data during reasoning.

Copied to clipboard

Zhen Li, Xiaohan Xu, Tao Shen, Can Xu, Jia-Chen Gu, Yuxuan Lai, Chongyang Tao, Shuai Ma

Visual Question Decomposition on Multimodal Large Language Models (2024.findings-emnlp)

Challenge:	introducing Large Language Models (LLMs) has opened new avenues for assessing generated content quality, e.g., coherence, creativity, and context relevance.
Approach:	They propose a taxonomy for organizing existing LLM-based evaluation metrics and a structured framework to understand and compare them.
Outcome:	The proposed taxonomy offers a framework to understand and compare LLM-based evaluation methods.

Copied to clipboard

Haowei Zhang, Jianzhe Liu, Zhen Han, Shuo Chen, Bailan He, Volker Tresp, Zhiqiang Xu, Jindong Gu

UNIKIE-BENCH: Benchmarking Large Multimodal Models for Key Information Extraction in Visual Documents (2026.acl-long)

Challenge:	Existing methods for question decomposition focus on unimodal language models, but question decomposing capability of Multimodal Large Language Models (MLLMs) has yet to be explored.
Approach:	They propose a finetuning dataset and a training objective for selective decomposition to enhance the model's question decomposing capability.
Outcome:	The proposed dataset shows that existing models struggle to produce high-quality sub-questions.

Copied to clipboard

Yifan Ji, Zhipeng Xu, Zhenghao Liu, Zulong Chen, Qian Zhang, Zhibo Yang, Junyang Lin, Yu Gu, Ge Yu, Maosong Sun

Information Extraction & Text Mining

Challenge:	Recent Large Multimodal Models (LMMs) have shown promising potential for performing end-to-end KIE directly from document images.
Approach:	They propose a benchmark to evaluate the performance of Large Multimodal Models (LMMs) using a constrained-category KIE track and an open-categorical KIE Track.
Outcome:	Experiments on 15 state-of-the-art LMMs show performance degradation under diverse schema definitions, long-tail key fields, and complex layouts, along with pronounced performance disparities across different document types and scenarios.

Detecting AI-Generated Video: A Vision–Language Dual-View Survey (2026.findings-acl)

Copied to clipboard

Dylan Xinming Hou, Juntian Zhang, Xu Gu, Yichen Wu, Nils Lukas, Gus Xia, Xiuying Chen, Yuhan Liu

HAD: HAllucination Detection Language Models Based on a Comprehensive Hallucination Taxonomy (2026.acl-industry)

Challenge:	realism of AI-generated Videos (AIGC-V) rendering artifact-centric detection insufficient, authors argue . a vision–language dual-view taxonomy is proposed to systematize this rapidly evolving field .
Approach:	They propose a Vision–Language Dual-View taxonomy to systematize AIGC-V detection . they propose realism of AI-generated Videos is rendering traditional inspection insufficient .
Outcome:	The proposed model aims to show that the existing methods are consistent with real-world facts.

Copied to clipboard

Fan Xu, Xinyu Hu, Zhenghan Yu, Li Lin, Xu Zhang, Yang Zhang, Wei Zhou, Jinjie Gu, Xiaojun Wan

TegTok: Augmenting Text Generation via Task-specific and Open-world Knowledge (2022.findings-acl)

Challenge:	relying on large language models for information has raised concerns about reliability and accuracy of outputs.
Approach:	They propose a hallucination taxonomy with 11 categories for various NLG tasks and propose HAllucination Detection models which integrate hallucinism detection, span-level identification, and correction into a single inference process.
Outcome:	The proposed models outperform baselines on HaluEval, FactCHD, and FaithBench, confirming their robustness and versatility.

Copied to clipboard

Chao-Hong Tan, Jia-Chen Gu, Chongyang Tao, Zhen-Hua Ling, Can Xu, Huang Hu, Xiubo Geng, Daxin Jiang

XMoE: Sparse Models with Fine-grained and Adaptive Expert Selection (2024.findings-acl)

Challenge:	Generating natural and informative texts has been a long-standing problem in NLP.
Approach:	They propose to augment TExt Generation via Task-specific and Open-world Knowledge in a unified framework.
Outcome:	The proposed model can learn what and how to generate on two text generation tasks.

Copied to clipboard

Yuanhang Yang, Shiyi Qi, Wenchao Gu, Chaozheng Wang, Cuiyun Gao, Zenglin Xu

Green & Sustainable NLP

Challenge:	XMoE leverages small experts and a threshold-based router to selectively engage only essential parameters.
Approach:	They propose a novel MoE that leverages small experts to selectively engage only essential parameters.
Outcome:	The proposed model can reduce computation load at MoE layers by over 50% without sacrificing performance.

Model Editing Harms General Abilities of Large Language Models: Regularization to the Rescue (2024.emnlp-main)

Copied to clipboard

Jia-Chen Gu, Hao-Xiang Xu, Jun-Yu Ma, Pan Lu, Zhen-Hua Ling, Kai-Wei Chang, Nanyun Peng

EcomScriptBench: A Multi-task Benchmark for E-commerce Script Planning via Step-wise Intention-Driven Product Association (2025.acl-long)

Challenge:	Existing methods that edit large language models with updated knowledge can cause side effects on the general abilities of LLMs such as reasoning, natural language inference, and question answering.
Approach:	They propose to regularize the edit update weights by imposing constraints on their complexity based on the RElative Change in weighT.
Outcome:	The proposed method can significantly mitigate the side effects while maintaining over 94% editing performance.

Copied to clipboard

Weiqi Wang, Limeng Cui, Xin Liu, Sreyashi Nag, Wenju Xu, Chen Luo, Sheikh Muhammad Sarwar, Yang Li, Hansu Gu, Hui Liu, Changlong Yu, Jiaxin Bai, Yifan Gao, Haiyang Zhang, Qi He, Shuiwang Ji, Yangqiu Song

EmoPrompt-ECPE: Emotion Knowledge-aware Prompt-tuning for Emotion-Cause Pair Extraction (2024.lrec-main)

Challenge:	Goal-oriented script planning is used by humans to plan for typical activities . however, this capability remains underexplored due to several challenges .
Approach:	They propose a framework that enables product-enriched scripts by associating products with each step based on the semantic similarity between the actions and their purchase intentions.
Outcome:	The proposed framework can generate product-enriched scripts from 2.4 million scripts . human annotations are conducted to provide gold labels for a sampled subset .

Copied to clipboard

Xue Gu, Zhihan Zhou, Ziyao Meng, Jian Li, Tiago Gomes, Adriano Tavares, Hao Xu

Low-Resource NLP

Challenge:	Existing methods for Emotion-cause pair extraction (ECPE) do not distinguish between the emotion-caused pairs that belong to different types of emotions, limiting their applicability.
Approach:	They propose an Emotion-cause pair extraction method which integrates the implicit knowledge of cause clauses into a prompt template and extends the emotion labels to categories with an external emotion word base.
Outcome:	The proposed method extracts all potential emotion clauses and corresponding cause clauses from unannotated documents.

QaRL: Rollout-Aligned Quantization-Aware RL for Fast and Stable Training under Training–Inference Mismatch (2026.findings-acl)

Copied to clipboard

Hao Gu, Hao Wang, Jiacheng Liu, Lujun Li, Qiyuan Zhu, Bei Liu, Binxing Xu, Lei Wang, Xintong Yang, Sida Lin, Sirui Han, Yike Guo

Challenge:	Recent work has shown that reinforcement learning with simple rule-based reward functions (RLVR) can induce emergent reasoning behaviors and yield gains in challenging domains such as math problem solving.
Approach:	They propose a rollout-alignment-quantization-aware RL which aligns training-side forward with the quantized rollout to minimize mismatch.
Outcome:	The proposed approach outperforms quantized-rollout training by +5.5 on Qwen3-30B-A3B MoE for math problems while maintaining low-bit throughput.

Lang2Act: Fine-Grained Visual Reasoning through Self-Emergent Linguistic Toolchains (2026.findings-acl)

Copied to clipboard

Yuqi Xiong, Chunyi Peng, Zhipeng Xu, Zhenghao Liu, Zulong Chen, Yukun Yan, Shuo Wang, Yu Gu, Ge Yu

BalanceSFT: Improving LLM Function Calling with Balanced Training Signals and Data Hardness (2026.findings-acl)

Challenge:	Existing frameworks depend on rigid, pre-defined external tools to extend perceptual capabilities of VLMs.
Approach:	They propose a framework that leverages self-emergent linguistic toolchains to enhance visual perception and reasoning.
Outcome:	The proposed framework improves the visual perception capabilities of large language models by incorporating external visual documents to address a given query.

Copied to clipboard

Bingguang Hao, Zengzhuang Xu, Maolin Wang, Yuntao Wen, Yicheng Chen, Cunyin Peng, Long Chen, Xiangyu Zhao, Jinjie Gu, Chenyi Zhuang, Ji Zhang

MIRAGE: Exploring How Large Language Models Perform in Complex Social Interactive Environments (2025.acl-short)

Challenge:	Currently, Supervised Fine-Tuning (SFT) is the prevailing method for equipping Large Language Models (LLMs) with function calling capabilities, but its effectiveness is often compromised by two challenges: 1) lengthy Chain-of-Thought (CoT) reasoning tokens dominate training signals over concise function calls in the learning objective; 2) scarcity of hard training examples.
Approach:	They propose a framework that uses a self-adjusted signal balancing loss and a hard data re-sampling strategy to selectively generate new, high-quality complex data guided by model errors.
Outcome:	The proposed framework surpasses state-of-the-art models like GPT-5 in function calling performance.

Copied to clipboard

Yin Cai, Zhouhong Gu, Zhaohan Du, Zheyu Ye, Shaosheng Cao, Yiqian Xu, Hongwei Feng, Ping Chen

EngiBench: A Benchmark for Evaluating Large Language Models on Engineering Problem Solving (2026.findings-acl)

Challenge:	Large Language Models (LLMs) have shown remarkable capabilities in environmental perception, reasoning-based decision-making, and simulating complex human behaviors, particularly in interactive role-playing contexts.
Approach:	They propose a framework to assess LLMs' proficiency in portraying advanced human behaviors through murder mystery games using eight intricately crafted scripts.
Outcome:	The framework evaluates LLMs' performance in portraying advanced human behaviors through murder mystery games.

Copied to clipboard

Xiyuan Zhou, Xinlei Wang, Yirui He, Ruixi Zou, Yang Wu, Yuheng Cheng, Yulu Xie, Wenxuan Liu, Huan Zhao, Yan Xu, Jinjin Gu, Junhua Zhao

Reasoning

Challenge:	Existing benchmarks focus on well-defined or abstract reasoning and fail to capture real-world engineering problems.
Approach:	They propose a hierarchical benchmark to evaluate large language models on engineering problems.
Outcome:	The proposed model performs well under well-defined conditions and is based on three levels of difficulty and covers diverse engineering subfields.

UEGP: Unified Expert-Guided Pre-training for Knowledge Rekindle (2024.findings-naacl)

Copied to clipboard

Yutao Mou, Kexiang Wang, Jianhe Lin, Dehong Ma, Jun Fan, Daiting Shi, Zhicong Cheng, Gu Simiu, Dawei Yin, Weiran Xu

GUI0: Self-Evolving Foundational GUI Agents in Super App Ecosystems (2026.acl-long)

Challenge:	Existing paradigms for pre-training and fine-tuning have limitations . knowledge rekindle aims to break through performance upper bounds of experts without introducing additional annotated data.
Approach:	They propose a new paradigm for pre-training and fine-tuning that aims to re-incorporate the fine- tuned expert model into the training cycle and break through performance upper bounds of experts.
Outcome:	The proposed model breaks through performance upper bounds of experts without additional annotated data.

Copied to clipboard

Xinyi Wang, Wei Dai, Kyle Qiao, Ke Wang, Peng Chen, Gang Cao, null Kangqin, Zhongpu Wang, Xiaode Zhang, Yanming Liu, Jihao Gu, Jingtao Xu, Gong Zhi

MPC-BERT: A Pre-Trained Language Model for Multi-Party Conversation Understanding (2021.acl-long)

Challenge:	Automated interaction with graphical user interfaces (GUIs) is central to general artificial intelligence, but remains challenging within Super App ecosystems.
Approach:	They propose a framework synergizing autonomous data synthesis with dual-agent co-evolution . GUI0 establishes a domain-aware foundation model via synthesized corpora and employs curriculum-driven reinforcement learning .
Outcome:	The proposed framework outperforms Gemini-2.5-Pro and Claude-4-Sonnet in the SuperAPP benchmark and has universal efficacy across base models.

Copied to clipboard

Jia-Chen Gu, Chongyang Tao, Zhenhua Ling, Can Xu, Xiubo Geng, Daxin Jiang

HSS-Synth: Humanities and Social Sciences Data Synthesis for LLMs (2026.findings-acl)

Challenge:	Existing models for multi-party conversation represent interlocutors and utterances individually . existing methods ignore complicated structure of MPC which may provide crucial interlocutor and tertiary semantics.
Approach:	They propose a pre-trained model for multi-party conversation that considers learning who says what to whom in a unified model with elaborated self-supervised tasks.
Outcome:	The proposed model outperforms existing models on three downstream tasks at two benchmarks.

Copied to clipboard

Ru Peng, Tianyu Zhao, Xijun Gu, Zhiting Fan, Haokai Xu, Jinyang Zhang, Yawen Zeng, Yihong Zhuang, Kexin Yang, Junyang Lin, Dayiheng Liu, Junbo Zhao

Challenge:	High-quality, diverse data are vital for large language models (LLMs) but remain scarce and costly.
Approach:	They define the first HSS domain system covering 14 mainstream fields and introduce HSS-Synth.
Outcome:	the proposed pipeline outperforms 14 leading baselines on 16 benchmarks.

Watermarking PLMs on Classification Tasks by Combining Contrastive Learning with Weight Perturbation (2023.findings-emnlp)

Copied to clipboard

Chenxi Gu, Xiaoqing Zheng, Jianhan Xu, Muling Wu, Cenyuan Zhang, Chengsong Huang, Hua Cai, Xuanjing Huang

Knowledge Mechanisms in Large Language Models: A Survey and Perspective (2024.findings-emnlp)

Challenge:	Large pre-trained language models (PLMs) are highly valuable intellectual property due to their expensive training costs.
Approach:	They propose to embed backdoors that can be triggered by specific inputs into models by model watermarking.
Outcome:	The proposed method can be used to protect the intellectual property of large pre-trained language models without knowledge about downstream tasks.

Copied to clipboard

Mengru Wang, Yunzhi Yao, Ziwen Xu, Shuofei Qiao, Shumin Deng, Peng Wang, Xiang Chen, Jia-Chen Gu, Yong Jiang, Pengjun Xie, Fei Huang, Huajun Chen, Ningyu Zhang

TMATH A Dataset for Evaluating Large Language Models in Generating Educational Hints for Math Word Problems (2025.coling-main)

Challenge:	Using large language models, we can understand knowledge mechanisms in LLMs for learning, storage, utilization, and evolution.
Approach:	They propose to analyze knowledge mechanisms in Large Language Models (LLMs) they examine utilization, evolution, and the potential dark knowledge (hypothesis) they hope to help understand knowledge in LLMs and provide insights for future research .
Outcome:	The proposed model can be used to analyze the evolution of parametric knowledge in LLMs.

Copied to clipboard

Changyong Qi, Yuang Wei, Haoxin Xu, Longwei Zheng, Peiji Chen, Xiaoqing Gu

Extending Context Window of Large Language Models from a Distributional Perspective (2024.emnlp-main)

Challenge:	Large Language Models (LLMs) are increasingly being applied in education, showing significant potential in personalized instruction, student feedback, and intelligent tutoring systems (ITSs).
Approach:	They propose a dataset specifically designed to evaluate LLMs’ ability to generate high-quality hints for Math Word Problems.
Outcome:	The proposed dataset shows that LLMs can generate more accurate and contextually appropriate educational hints for math word problems without offering direct answers.

Copied to clipboard

Yingsheng Wu, Yuxuan Gu, Xiaocheng Feng, Weihong Zhong, Dongliang Xu, Qing Yang, Hongtao Liu, Bing Qin

CPRM: A LLM-based Continual Pre-training Framework for Relevance Modeling in Commercial Search (2025.naacl-industry)

Challenge:	Existing scaling methods for extending context window rely on empirical approaches and lack understanding of the internal distribution within RoPE resulting in suboptimal performance.
Approach:	They propose to optimize the context window extending task from the view of rotary angle distribution by minimizing disturbance between rotary angles to maintain consistency with the pre-training phase.
Outcome:	The proposed approach reduces by up to 72% of the distributional disturbance when extending LLaMA2’s context window to 8k, and reduces it by up 32% when extending to 16k.

Copied to clipboard

Kaixin Wu, Yixin Ji, Zeyuan Chen, Qiang Wang, Cunxiang Wang, Hong Liu, Baijun Ji, Xu Jia, Zhongyi Liu, Jinjie Gu, Yuan Zhou, Linjian Mo

RLShield: Dynamic Jailbreak Detection for LLMs via Reinforced Adaptive Learning (2026.findings-acl)

Challenge:	Relevance modeling between queries and items is a key component of commercial search engines.
Approach:	They propose a framework for continual pre-training of LLMs to enhance domain knowledge . they employ queries and multi-field item to jointly pre-train for enhancing domain knowledge.
Outcome:	The proposed model achieves convincing performance compared to strong baselines.

Copied to clipboard

Zhao Tong, Pengfei Yang, Yimeng Gu, Haichao Shi, Qiang Liu, Xingcheng Xu, Shu Wu, Xiao-Yu Zhang