Papers by Song Gao

Topology Imbalance and Relation Inauthenticity Aware Hierarchical Graph Attention Networks for Fake News Detection (2022.coling-1)

Challenge:	Existing methods focus on replicating dialogues in textual form, neglecting the role’s voice traits as a crucial effect in interaction, which tends to be more immersive experiences in realistic scenarios.
Approach:	They propose a first seamless speech-language personality interaction model to achieve immersive RPAs with low latency.
Outcome:	The proposed model exhibits role-specific personality traits and vocal traits throughout the interaction, enabling a mixture of speech and language responses.

Copied to clipboard

Li Gao, Lingyun Song, Jie Liu, Bolin Chen, Xuequn Shang

EcoSafeRAG: Efficient Security through Context Analysis in Retrieval-Augmented Generation (2025.findings-emnlp)

Challenge:	Existing methods to detect fake news focus on mining lexical and syntactic features.
Approach:	They propose a topology imbalance and Relation inauthenticity aware Hierarchical Graph Attention Networks to identify fake news on social media.
Outcome:	The proposed method outperforms state-of-the-art methods on real-world datasets.

Copied to clipboard

Ruobing Yao, Yifei Zhang, Shuang Song, Neng Gao, Chenyang Tu

Green & Sustainable NLP

Challenge:	Existing defense methods rely on internal knowledge of the model, which conflicts with the design concept of Retrieval-Augmented Generation (RAG).
Approach:	EcoSafeRAG uses sentence-level processing and bait-guided context diversity detection to identify malicious content .
Outcome:	EcoSafeRAG uses sentence-level processing and bait-guided context diversity detection to identify malicious content.

PhysicsArena: The First Multimodal Physics Reasoning Benchmark Exploring Variable, Process, and Solution Dimensions (2025.findings-emnlp)

Copied to clipboard

Song Dai, Yibo Yan, Jiamin Su, Zihao Dongfang, Yubo Gao, Yonghua Hei, Jungang Li, Junyan Zhang, Sicheng Tao, Zhuoran Gao, Xuming Hu

RP-DNN: A Tweet Level Propagation Context Based Deep Neural Networks for Early Rumor Detection in Social Media (2020.lrec-1)

Challenge:	Current physics benchmarks focus on text-only inputs or only on problem-solving . current physics reasoning benchmarks neglect critical intermediate steps of variable identification and process formulation.
Approach:	a new benchmark evaluates multimodal large language models in physics reasoning . the benchmark measures variables, process formulations, and solution derivation .
Outcome:	PhysicsArena is the first multimodal physics reasoning benchmark . it evaluates MLLMs across three critical dimensions: variable identification, process formulation, and solution derivation.

Copied to clipboard

Jie Gao, Sooji Han, Xingyi Song, Fabio Ciravegna

AffectCodec: Emotion-Preserving Neural Speech Codec for Expressive Speech Modeling (2026.findings-acl)

Challenge:	Existing methods for early rumor detection on social media platforms are limited, incomplete and noisy.
Approach:	They propose a novel hybrid neural network architecture which combines a task-specific character-based bidirectional language model and stacked Long Short-Term Memory (LSTM) networks to represent textual contents and social-temporal contexts of input source tweets.
Outcome:	The proposed model achieves state-of-the-art for detecting unseen rumors on large augmented data which covers more than 12 events and 2,967 rumors.

Copied to clipboard

Jiacheng Shi, Hongfei Du, Xinyuan Song, Y. Alicia Hong, Yanfu Zhang, Ashley Gao

Towards the Law of Capacity Gap in Distilling Language Models (2025.acl-long)

Challenge:	Existing codecs optimize acoustic reconstruction, leaving emotion expressiveness insufficiently modeled at the representation level.
Approach:	They propose an emotion-guided neural speech codec that preserves emotional information while maintaining semantic fidelity and prosodic naturalness.
Outcome:	The proposed codec preserves emotional cues while maintaining semantic fidelity and prosodic naturalness.

Copied to clipboard

Chen Zhang, Qiuchi Li, Dawei Song, Zheyu Ye, Yan Gao, Yao Hu

ServImage: An Image Generation and Editing Benchmark from Real-world Commercial Imaging Services (2026.acl-long)

Challenge:	Language model (LM) distillation aims at distilling knowledge in a large teacher LM to a small student one.
Approach:	They propose to use the law of capacity gap to distill knowledge from a large teacher to a small student model.
Outcome:	The proposed model outperforms other language models on a larger scale by using the law of capacity gap inducted from a preliminary study on small-scale (3B) LMs.

Copied to clipboard

Fengxian Ji, Jingpu Yang, Zirui Song, Lang Gao, Junhong Liang, Zhenhao Chen, Jinghui Zhang, Xiuying Chen

Iterative Document Representation Learning Towards Summarization with Polishing (D18-1)

Challenge:	Recent image generation and editing models demonstrate robust adherence to instructions and high visual quality on academic benchmarks.
Approach:	They propose a benchmark that correlates image outputs with economic value in commercial design projects.
Outcome:	ServImage benchmarks show image generation models perform well on academic benchmarks but are uncertain on commercial projects.

Copied to clipboard

Xiuying Chen, Shen Gao, Chongyang Tao, Yan Song, Dongyan Zhao, Rui Yan

Information Extraction & Text Mining

Challenge:	Existing summarization methods read through document only once to generate a document representation, resulting in a sub-optimal representation.
Approach:	They propose an iterative model for supervised extractive text summarization which polishes the document representation on many passes through the document.
Outcome:	The proposed model outperforms state-of-the-art extractive systems on CNN/DailyMail and DUC2002 datasets.

Z-Code++: A Pre-trained Language Model Optimized for Abstractive Summarization (2023.acl-long)

Copied to clipboard

Pengcheng He, Baolin Peng, Song Wang, Yang Liu, Ruochen Xu, Hany Hassan, Yu Shi, Chenguang Zhu, Wayne Xiong, Michael Zeng, Jianfeng Gao, Xuedong Huang

Confront Insider Threat: Precise Anomaly Detection in Behavior Logs Based on LLM Fine-Tuning (2025.coling-main)

Challenge:	Z-Code++ is a pre-trained language model optimized for abstractive text summarization.
Approach:	They propose a pre-trained language model optimized for abstractive text summarization that uses a two-phase pre-training technique to improve model's performance.
Outcome:	The proposed model outperforms the competing models on low-resource summarization tasks in zero-shot and few-shot settings.

Copied to clipboard

Shuang Song, Yifei Zhang, Neng Gao

DataSeer: A Manager-Centric Collaborative Multi-Agent Framework with Multi-Branch Reasoning for Automated Insight Discovery (2026.findings-acl)

Challenge:	Current methods for insider threat detection suffer from low precision and information loss . a novel approach to detect insider threats is needed to improve accuracy .
Approach:	They propose a precise anomaly detection solution based on Large Language Model (LLM) fine-tuning . they represent user behavior in natural language and implement a threat tracing mechanism .
Outcome:	The proposed solution achieves an F1 score of 0.8941 on the CERT v6.2 dataset .

Copied to clipboard

Suchen Liu, Yuanfeng Song, Jun Gao, Xing Chen

MMEvol: Empowering Multimodal Large Language Models with Evol-Instruct (2025.findings-acl)

Challenge:	Existing methods for automated insight discovery lack contextual coherence and coverage due to single-path exploration.
Approach:	They propose a Manager-Centric Collaborative Framework that integrates planner and executor . it ensures cross-episode contextual coherence and allows for adaptive sub-goal generation .
Outcome:	The proposed framework outperforms baselines on InsightBench and Inseval.

Copied to clipboard

Run Luo, Haonan Zhang, Longze Chen, Ting-En Lin, Xiong Liu, Yuchuan Wu, Min Yang, Yongbin Li, Minzheng Wang, Pengpeng Zeng, Lianli Gao, Heng Tao Shen, Yunshui Li, Hamid Alinejad-Rokny, Xiaobo Xia, Jingkuan Song, Fei Huang

Under the Shadow of Babel: How Language Shapes Reasoning in LLMs (2025.findings-emnlp)

Challenge:	a new framework for image-text instruction data evolution improves MLLM performance . lack of high-quality instruction data remains a major bottleneck in ML modeling .
Approach:	They propose a multimodal instruction data evolution framework that iteratively enhances data quality through fine-grained perception, cognitive reasoning, and interaction evolution.
Outcome:	The proposed approach improves MLLM performance in nine vision-language tasks while using significantly less data.

Copied to clipboard

Chenxi Wang, Yixuan Zhang, Lang Gao, Zixiang Xu, Zirui Song, Yanbo Wang, Xiuying Chen

Linguistics & Cognitive NLP

Challenge:	linguistic relativity suggests that the structure of language shapes cognitive patterns . large language models internalize the habitual logical structures embedded in different languages, authors say .
Approach:	a study introduces a bilingual dataset for causal reasoning in Chinese and English.
Outcome:	a new study shows that large language models internalize reasoning biases shaped by language . the model internalizes language-specific preferences and rigidly applies them to atypical inputs, the study shows .

Audio Jailbreak: An Open Comprehensive Benchmark for Jailbreaking Large Audio-Language Models (2026.acl-long)

Copied to clipboard

Zirui Song, Qian Jiang, Mingxuan Cui, Mingzhe Li, Lang Gao, Zeyu Zhang, Zixiang Xu, Yanbo Wang, Guangxian Ouyang, Zhenhao Chen, Xiuying Chen

PACE: Prefix-Protected and Difficulty-Aware Compression for Efficient Reasoning (2026.findings-acl)

Challenge:	a recent study evaluated large audio-language models against jailbreak attacks . a new benchmark is being developed to evaluate LAM safety against jailbreaking attacks based on temporal and semantic nature of speech .
Approach:	They propose a benchmark to evaluate LAM jailbreak vulnerabilities in adversarial audio prompts . they use a dataset of 1,495 adversarials to evaluate their performance .
Outcome:	The proposed benchmark evaluates state-of-the-art LAMs against jailbreak attacks . it demonstrates that even small, semantically preserved perturbations can reduce safety .

Copied to clipboard

Ruixiang Feng, Yuntao Wen, Silin Zhou, Ke Shi, Yifan Wang, Ran Le, Zhenwei An, Zongchao Chen, Chen Yang, Guangyue Peng, Yiming Jia, Dongsheng Wang, Tao Zhang, Lisi Chen, Yang Song, Shen Gao, Shuo Shang

Model Surgery: Modulating LLM’s Behavior Via Simple Parameter Editing (2025.naacl-long)

Challenge:	Existing LRMs often suffer from "overthinking" and excessively long reasoning traces . a dual-level framework for length compression of LRM is proposed .
Approach:	They propose a framework for prefix-protected and difficulty-aware compression under hierarchical supervision.
Outcome:	The proposed framework reduces token usage while improving accuracy on math benchmarks.

Copied to clipboard

Huanqian Wang, Yang Yue, Rui Lu, Jingxin Shi, Andrew Zhao, Shenzhi Wang, Shiji Song, Gao Huang

Challenge:	Current approaches for detoxification or preventing jailbreaking involve fine-tuning billions of parameters through gradient descent with substantial computational cost.
Approach:	They propose to use supervised fine-tuning and Reinforcement Learning from human feedback to modify LLMs' behavior by directly editing a small subset of parameters.
Outcome:	Experiments show that editing a small subset of parameters can modulate specific behaviors of LLMs, such as detoxification and resistance to jailbreak, with only inference-level computational resources.

Hallucinations as Orthogonal Noise: Inference-Time Manifold Alignment via Dynamic Contextual Orthogonalization (2026.findings-acl)

Copied to clipboard

Mingkuan Zhao, Wentao Hu, Tianchen Huang, Yuheng Min, Suquan Chen, Yide Gao, Yanbo Zhai, Shuangyong Song, Xuelong Li

Using Customer Service Dialogues for Satisfaction Analysis with Context-Assisted Multiple Instance Learning (D19-1)

Challenge:	Hallucinations in Large Language Models persist in critical domains where generated content diverges from contextual facts or logical constraints.
Approach:	They propose to generate hallucinations as orthogonal noise relative to the semantic manifold of the residual stream.
Outcome:	The proposed method achieves superior contextual faithfulness compared to state-of-the-art methods.

Copied to clipboard

Kaisong Song, Lidong Bing, Wei Gao, Jun Lin, Lujun Zhao, Jiancheng Wang, Changlong Sun, Xiaozhong Liu, Qiong Zhang

Natural Language Interfaces

Challenge:	Existing studies fail to provide comprehensive service satisfaction analysis . Existing models fail to include satisfaction polarity classification and sentimental utterance identification .
Approach:	They propose a model that predicts customer sentiments and aggregates them into service satisfaction polarity.
Outcome:	The proposed model predicts customer sentiments and aggregates them into service satisfaction polarity and reasoning clues.

Multilingual Generative Retrieval via Cross-lingual Semantic Compression (2025.findings-emnlp)

Copied to clipboard

Yuxin Huang, Simeng Wu, Ran Song, Yan Xiang, Yantuan Xian, Shengxiang Gao, Zhengtao Yu

Multilinguality

Challenge:	Existing methods for multilingual retrieval still face cross-lingual identifier misalignment and identifiere inflation.
Approach:	They propose a framework that unifies semantically equivalent multilingual keywords into shared atoms to align semantics and compresses the identifier space.
Outcome:	The proposed framework improves cross-lingual alignment and reduces redundancy.

High-Order Semantic Alignment for Unsupervised Fine-Grained Image-Text Retrieval (2024.lrec-main)

Copied to clipboard

Rui Gao, Miaomiao Cheng, Xu Han, Wei Song

ParetoRAG: Leveraging Sentence-Context Attention for Robust and Efficient Retrieval-Augmented Generation (2025.findings-emnlp)

Challenge:	Existing studies focus on learning global or local correspondence, but lack fine-grained local-global alignment.
Approach:	They propose a High Order Semantic Alignment (HOSA) model that can provide complementary and comprehensive semantic clues to infer correlation scores.
Outcome:	The proposed model outperforms state-of-the-art models in retrieving the most relevant results.

Copied to clipboard

Ruobing Yao, Yifei Zhang, Shuang Song, Yuhan Liu, Neng Gao, Chenyang Tu

SynFix: Dependency-Aware Program Repair via RelationGraph Analysis (2025.findings-acl)

Challenge:	Retrieval-augmented generation systems face persistent challenges in retrieval inefficiency and the inability of LLMs to filter out irrelevant information.
Approach:	They propose an unsupervised framework that optimizes RAG systems through sentence-level refinement guided by the Pareto principle.
Outcome:	The proposed framework achieves dual improvements in retrieval precision and generation quality without additional training or API resources while using only 40% of the tokens compared to traditional approaches.

Copied to clipboard

Xunzhu Tang, Jiechao Gao, Jin Xu, Tiezhu Sun, Yewei Song, Saad Ezzini, Wendkûuni C. Ouédraogo, Jacques Klein, Tegawendé F. Bissyandé

Decoupling Mixture-of-Graphs: Unseen Relational Learning for Knowledge Graph Completion by Fusing Ontology and Textual Experts (2022.coling-1)

Challenge:	Existing methods for resolving repository-level debugging are limited by their interdependencies.
Approach:	They propose a RelationGraph-based approach that integrates large language models with structural search and synchronization techniques for coordinated program repair across codebases.
Outcome:	SynFix resolves 52.33% of issues in SWE-bench-lite, 55.8% in Swe-bech-verified and 29.86% in S WE-beach-full.

Copied to clipboard

Ran Song, Shizhu He, Suncong Zheng, Shengxiang Gao, Kang Liu, Zhengtao Yu, Jun Zhao

MLLM-Bench: Evaluating Multimodal LLMs with Per-sample Criteria (2025.naacl-long)

Challenge:	Existing methods for knowledge Graph Completion (KGC) fail in unseen relation representations.
Approach:	They propose to use three kinds of graphs to obtain unseen relation representations . they propose to decouple mixture-of-graph experts (DMoG) for unseened relations learning .
Outcome:	The proposed method outperforms the state-of-the-art methods on unseen relation representations.

Copied to clipboard

Wentao Ge, Shunian Chen, Hardy Chen, Nuo Chen, Junying Chen, Zhihong Chen, Wenya Xie, Shuo Yan, ChenghaoZhu ChenghaoZhu, Ziyue Lin, Dingjie Song, Xidong Wang, Anningzhe Gao, Zhang Zhiyi, Jianquan Li, Xiang Wan, Benyou Wang

MoDification: Mixture of Depths Made Easy (2025.naacl-long)

Challenge:	Existing evaluation methodologies for multimodal large language models are limited in evaluating objective queries without considering real-world user experiences.
Approach:	They propose to evaluate multimodal large language models with per-sample criteria using potent MLLM as the judge.
Outcome:	The proposed evaluation paradigm shows that it can be used to evaluate multimodal large language models with per-sample criteria.

Copied to clipboard

Chen Zhang, Meizhi Zhong, Qimeng Wang, Xuantao Lu, Zheyu Ye, Chengqiang Lu, Yan Gao, Yao Hu, Kehai Chen, Min Zhang, Dawei Song

Green & Sustainable NLP

Challenge:	Long-context efficiency is a trending topic in large language model (LLM) serving.
Approach:	They propose a method to combine long-context efficiency and mixture of depths to bring down both latency and memory.
Outcome:	The proposed method achieves 1.2 speedup in latency and 1.8 reduction in memory compared to original LLMs especially in long-context applications.

Rethink Rumor Detection in the Era of LLMs: A Review (2025.findings-emnlp)

Copied to clipboard

Chang Yang, Peng Zhang, Jing Zhang, Hui Gao, Changhao Song

Mixture of Weight-shared Heterogeneous Group Attention Experts for Dynamic Token-wise KV Optimization (2025.emnlp-main)

Challenge:	rumor detection has been reshaped by large language models (LLMs) this paper proposes a Cognition-Interaction-Behavior (CIB) framework for rumour detection based on collective intelligence .
Approach:	They propose a Cognition-Interaction-Behavior framework for rumor detection based on collective intelligence and explore synergistic relationship between LLMs and collective intelligence in rumour governance.
Outcome:	The proposed framework unifies existing methods and reveals synergistic relationship between LLMs and collective intelligence in rumor governance.

Copied to clipboard

Guanghui Song, Dongping Liao, Yiren Zhao, Kejiang Ye, Cheng-zhong Xu, Xitong Gao

Green & Sustainable NLP

Challenge:	Existing methods for token-level KV optimization and grouping of tokens are inefficient and strain compute and storage resources.
Approach:	They propose a mixture-of-expert approach that dynamically optimizes token-wise computation and memory allocation by a token-based expert-choice routing mechanism guided by learned importance scores.
Outcome:	The proposed approach retains all tokens while adaptively routing them to specialized experts with varying KV group sizes, balancing granularity and efficiency.

Does Large Language Model Contain Task-Specific Neurons? (2024.emnlp-main)

Copied to clipboard

Ran Song, Shizhu He, Shuting Jiang, Yantuan Xian, Shengxiang Gao, Kang Liu, Zhengtao Yu

ReAgent: Reversible Multi-Agent Reasoning for Knowledge-Enhanced Multi-Hop QA (2025.emnlp-main)

Challenge:	Large language models (LLMs) have demonstrated remarkable capabilities in comprehensively handling various types of natural language processing (NLP) tasks.
Approach:	They propose a method for task-specific neuron localization based on Causal Gradient Variation with Special Tokens (CGVST) this method identifies task- specific neurons by concentrating on the most significant tokens during task processing, eliminating redundant tokens and minimizing interference from non-essential neurons.
Outcome:	The proposed method can locate task-specific neurons across eight public tasks.

Copied to clipboard

Zhao Xinjie, Fan Gao, Xingyu Song, Yingjian Chen, Rui Yang, Yanran Fu, Yuyang Wang, Yusuke Iwasawa, Yutaka Matsuo, Irene Li

Debiased Orthogonal Boundary-Driven Efficient Noise Mitigation (2026.acl-long)

Challenge:	Multi-hop question answering (QA) is a central challenge in natural language processing . early mistakes can cause errors and undermine the final result, authors say .
Approach:	They propose a reversible multi-agent reasoning framework that backtracks to earlier valid states when conflicts arise.
Outcome:	Empirical evaluation shows that the framework improves on forward-only benchmarks by 6% . the approach enables agents to backtrack to valid states when conflicts arise .

Copied to clipboard

Hao Li, Jiayang Gu, Jingkuan Song, An Zhang, Lianli Gao

LLM-SLM Collaborative Framework of Idiomatic Expression Generation (2026.acl-long)

Challenge:	Existing noise mitigation methods face limitations due to their task-specific design, model dependency, and significant computational overhead.
Approach:	They propose a model-agnostic noise label mitigation paradigm that uses an estimator model and a scoring function to assess the noise level of input pairs.
Outcome:	The proposed method is superior to existing methods in training tasks and tasks.

Copied to clipboard

Hui Gao, Changhao Song, Peng Zhang, Jing Zhang, Chang Yang, Liuxian Ge

Model-Generated Pretraining Signals Improves Zero-Shot Generalization of Text-to-Text Transformers (2023.acl-long)

Challenge:	Existing methods for idiomatic expression generation lack parallel data and manual annotations.
Approach:	They propose an iterative LLM-SLM collaborative framework that replaces human supervision for idiomatic expression data generation.
Outcome:	The proposed framework outperforms DeepSeek-R1 in Chinese Idiom Polishing with a 25.2% improvement in accuracy.

Copied to clipboard

Linyuan Gong, Chenyan Xiong, Xiaodong Liu, Payal Bajaj, Yiqing Xie, Alvin Cheung, Jianfeng Gao, Xia Song

Breaking Consensus Bias: Unsupervised Reinforcement Learning for Machine Translation (2026.findings-acl)

Challenge:	Recent work in NLP has shown that pretrained language models have made notable progress toward generalization to unseen tasks.
Approach:	They propose to pretrain T5 using an auxiliary model to construct more challenging token replacements for the main model to denoise.
Outcome:	The proposed model outperforms similar-sized baseline models on prompted NLP benchmarks and rivals the state-of-the-art model with only 8% of its parameters.

Copied to clipboard

Shuting Jiang, Ran Song, Siqi Zhang, Yuxin Huang, Shengxiang Gao, Zhengtao Yu

Machine Translation

Challenge:	Existing RL approaches for MT face fixed references or the production of homogeneous references leading to mode collapse in unsupervised settings.
Approach:	They propose an Entropy-Driven Unsupervised RL framework for machine translation that leverages entropy for supervision construction and self-evolution.
Outcome:	The proposed framework outperforms supervised and unsupervised baselines in multiple language pairs.

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

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

When Personalization Tricks Detectors: The Feature-Inversion Trap in Machine-Generated Text Detection (2026.acl-long)

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

Lang Gao, Xuhui Li, Chenxi Wang, Mingzhe Li, Wei Liu, Zirui Song, Jinghui Zhang, Rui Yan, Preslav Nakov, Xiuying Chen

Robustness in NLP

Challenge:	Personalized MGT detection remains largely underexplored due to personalization challenges . large language models (LLMs) can imitate personal writing styles, but they can generate fake news and misinformation.
Approach:	They propose a benchmark to evaluate detector robustness under personalization . they attribute this limitation to a feature-inversion trap that flips the effect in personalized contexts .
Outcome:	The proposed framework predicts detector robustness under personalization with an 85% correlation to actual results.

Topic Memory Networks for Short Text Classification (D18-1)

Copied to clipboard

Jichuan Zeng, Jing Li, Yan Song, Cuiyun Gao, Michael R. Lyu, Irwin King

Information Extraction & Text Mining

Challenge:	Existing classification models for short texts are weak due to data sparsity .
Approach:	They propose topic memory networks for short text classification with a novel topic memory mechanism to encode latent topic representations indicative of class labels.
Outcome:	The proposed model outperforms state-of-the-art models on short text classification, while generating coherent topics.

GAST: Gradient-aligned Sparse Tuning of Large Language Models with Data-layer Selection (2026.eacl-long)

Copied to clipboard

Kai Yao, Zhenghan Song, Kaixin Wu, Mingjie Zhong, Danzhao Cheng, Zhaorui Tan, Yixin Ji, Penglei Gao

Boosting LLM Agents with Recursive Contemplation for Effective Deception Handling (2024.findings-acl)

Challenge:	Existing methods focus on layer-selective and data-selectory fine-tuning, but ignore the fact that different data points contribute varying degrees to distinct model layers.
Approach:	They propose a method that performs selective fine-tuning at both data and layer dimensions as integral components of a unified optimization strategy.
Outcome:	Experiments show that the proposed method outperforms baseline methods in terms of performance and performance.

Copied to clipboard

Shenzhi Wang, Chang Liu, Zilong Zheng, Siyuan Qi, Shuo Chen, Qisen Yang, Andrew Zhao, Chaofei Wang, Shiji Song, Gao Huang

Challenge:	Recent advances in large language models (LLMs) have led to significant success in using LLMs as agents.
Approach:	They propose a cognitive framework that incorporates first-order and second-order perspective transitions into LLMs to enhance their ability to identify and counteract deceptive information.
Outcome:	The proposed framework enhances LLMs’ ability to identify and counteract deceptive information without extra fine-tuning and data.

Unlocking Smarter Device Control: Foresighted Planning with a World Model-Driven Code Execution Approach (2025.findings-emnlp)

Copied to clipboard

Xiaoran Yin, Xu Luo, Hao Wu, Lianli Gao, Jingkuan Song

Challenge:	Current approaches to automating complex tasks focus on reactive policies and focus on visual observations.
Approach:	They propose a framework that prioritizes natural language understanding and structured reasoning to enhance the agent’s global understanding of the environment by developing a task-oriented, refinable world model at the outset of the task.
Outcome:	The proposed framework outperforms existing approaches in simulated environments and on real mobile devices.

IRT-Router: Effective and Interpretable Multi-LLM Routing via Item Response Theory (2025.acl-long)

Copied to clipboard

Wei Song, Zhenya Huang, Cheng Cheng, Weibo Gao, Bihan Xu, GuanHao Zhao, Fei Wang, Runze Wu

Linguistic Theories

Challenge:	Large language models have demonstrated exceptional performance across a wide range of tasks . however, selecting the optimal LLM to respond to a user query often necessitates a delicate balance between performance and cost.
Approach:	They propose a multi-LLM routing framework that efficiently routes user queries to the most suitable LLM.
Outcome:	The proposed framework outperforms baseline methods in terms of effectiveness and interpretability.

OpenCoder: The Open Cookbook for Top-Tier Code Large Language Models (2025.acl-long)

Copied to clipboard

Siming Huang, Tianhao Cheng, Jason Klein Liu, Weidi Xu, Jiaran Hao, Liuyihan Song, Yang Xu, Jian Yang, Jiaheng Liu, Chenchen Zhang, Linzheng Chai, Ruifeng Yuan, Xianzhen Luo, Qiufeng Wang, YuanTao Fan, Qingfu Zhu, Zhaoxiang Zhang, Yang Gao, Jie Fu, Qian Liu, Houyi Li, Ge Zhang, Yuan Qi, Xu Yinghui, Wei Chu, Zili Wang

LongDocURL: a Comprehensive Multimodal Long Document Benchmark Integrating Understanding, Reasoning, and Locating (2025.acl-long)

Challenge:	Code LLMs lack reproducible data pipelines and training protocols for reproducible advancements in code intelligence.
Approach:	They propose a top-tier code LLM that releases model weights and inference code . reproducible data pipelines, rigorous experimental ablation results and training protocols are included .
Outcome:	The proposed model achieves comparable performance to leading models and serves as an "open cookbook" reproducible training data, rigorous experimental ablation results, and detailed training protocols are also included in the model.

Copied to clipboard

Chao Deng, Jiale Yuan, Pi Bu, Peijie Wang, Zhong-Zhi Li, Jian Xu, Xiao-Hui Li, Yuan Gao, Jun Song, Bo Zheng, Cheng-Lin Liu

Word Form Matters: LLMs’ Semantic Reconstruction under Typoglycemia (2025.findings-acl)

Challenge:	Existing document understanding benchmarks only handle a small number of pages . existing models are limited to handling only a limited number of documents .
Approach:	They propose a long document understanding benchmark that integrates three primary tasks and 20 sub-tasks based on different primary tasks.
Outcome:	The proposed model outperforms existing benchmarks on open-source and closed-source models . the model outpersforms other models on more than 33,000 pages of documents .

Copied to clipboard

Chenxi Wang, Tianle Gu, Zhongyu Wei, Lang Gao, Zirui Song, Xiuying Chen

Challenge:	Typoglycemia is a phenomenon where people can read words even when the middle letters of the words are scrambled.
Approach:	They propose a reliable metric to quantify the degree of semantic reconstruction and validate its effectiveness.
Outcome:	The proposed metric quantifies the degree of semantic reconstruction and validates its effectiveness.

AutoSDT: Scaling Data-Driven Discovery Tasks Toward Open Co-Scientists (2025.emnlp-main)

Copied to clipboard

Yifei Li, Hanane Nour Moussa, Ziru Chen, Shijie Chen, Botao Yu, Mingyi Xue, Benjamin Burns, Tzu-Yao Chiu, Vishal Dey, Zitong Lu, Chen Wei, Qianheng Zhang, Tianyu Zhang, Song Gao, Xuhui Huang, Xia Ning, Nesreen K. Ahmed, Ali Payani, Huan Sun

Towards A Better Initial Policy Model For Scalable Long-CoT Reinforcement Learning (2025.findings-acl)

Challenge:	AutoSDT-5K is the only automatically collected and the largest open dataset for data-driven scientific discovery.
Approach:	They propose an automatic pipeline that collects high-quality coding tasks in real-world data-driven discovery workflows.
Outcome:	The proposed pipeline synthesizes accurate tasks and tasks from a dataset of 5,404 tasks covering four scientific disciplines and 756 Python packages.

Copied to clipboard

Bofei Gao, Yejie Wang, Yibo Miao, Ruoyu Wu, Feifan Song, Longhui Yu, Tianyu Liu, Baobao Chang

Lock on Target! Precision Unlearning via Directional Control (2025.findings-emnlp)

Challenge:	Long-CoT reasoning and reinforcement learning are demonstrating remarkable performance and scalability, however, there is a lack of systematic guidelines for obtaining a better initial policy model.
Approach:	They propose a systematic guideline and a novel Re-RFT method to obtain more efficient reasoning patterns from different initial models.
Outcome:	The proposed method surpasses DeepSeek-R1-Distill-Qwen-14B model by 4.6%, demonstrating its effectiveness and superiority.

Copied to clipboard

Yuntao Wen, Ruixiang Feng, Feng Guo, Yifan Wang, Ran Le, Yang Song, Shen Gao, Shuo Shang

Challenge:	Existing methods for unlearning harmful, sensitive, or outdated knowledge suffer from two critical limitations: (1) collateral forgetting, where erasing target data inadvertently removes related but desirable knowledge, and (2) generality forgetting degrades the model’s general capabilities.
Approach:	They propose a method that identifies and leverages a targeted "unlearning direction" in the model's parameter space and selectively updates along this direction.
Outcome:	Experiments show that the proposed method achieves state-of-the-art unlearning precision while preserving both related knowledge and general capabilities.

Multilingual Knowledge Graph Completion from Pretrained Language Models with Knowledge Constraints (2023.findings-acl)

Copied to clipboard

Ran Song, Shizhu He, Shengxiang Gao, Li Cai, Kang Liu, Zhengtao Yu, Jun Zhao

Harnessing the Power of Large Language Model for Uncertainty Aware Graph Processing (2024.lrec-main)

Challenge:	Existing methods for multilingual knowledge graph completion do not align with mKGC tasks because of their English-centric bias.
Approach:	They propose to use multilingual pretrained language models to solve queries in different languages by reasoning a tail entity.
Outcome:	The proposed method outperforms the previous SOTA on Hits@1 and Hits @10 by 12.32% and 16.03% on public datasets.

Copied to clipboard

Zhenyu Qian, Yiming Qian, Yuting Song, Fei Gao, Hai Jin, Chen Yu, Xia Xie

SessionIntentBench: A Multi-task Inter-session Intention-shift Modeling Benchmark for E-commerce Customer Behavior Understanding (2026.findings-acl)

Challenge:	Existing methods for graph processing rely on assumptions about data relations that are inadequate when handling large and complex graph data.
Approach:	They propose a large language model enhanced by an uncertainty-aware module to provide a confidence score on the generated graph data.
Outcome:	The proposed approach surpasses state-of-the-art algorithms by a substantial margin on ten datasets.

Copied to clipboard

Yuqi Yang, Weiqi Wang, Baixuan Xu, Wei Fan, Qing Zong, Chunkit Chan, Zheye Deng, Xin Liu, Yifan Gao, Changlong Yu, Chen Luo, Yang Li, Zheng Li, Qingyu Yin, Bing Yin, Yangqiu Song

FolkScope: Intention Knowledge Graph Construction for E-commerce Commonsense Discovery (2023.findings-acl)

Challenge:	Existing models fail to capture and model customer intention effectively because of insufficient information exploitation and only apparent information like descriptions and titles are used.
Approach:	They propose to exploit existing session data to capture and model intention in E-commerce product purchase sessions using a multimodal benchmark.
Outcome:	The proposed framework can bridge the gap between intention understanding in simplified research cases like co-buy intention and more complex yet practical scenarios like session history.

Copied to clipboard

Changlong Yu, Weiqi Wang, Xin Liu, Jiaxin Bai, Yangqiu Song, Zheng Li, Yifan Gao, Tianyu Cao, Bing Yin

Commonsense Reasoning

Challenge:	Existing intention-based studies on recommendation tasks are limited and use models to implicitly model the intention memberships.
Approach:	They propose a framework that leverages the generation power of large language models and human-in-the-loop annotation to semi-automatically construct the intention knowledge graph.
Outcome:	The proposed framework can model e-commerce knowledge and have many potential applications.

GazeVQA: A Video Question Answering Dataset for Multiview Eye-Gaze Task-Oriented Collaborations (2023.emnlp-main)

Copied to clipboard

Muhammet Ilaslan, Chenan Song, Joya Chen, Difei Gao, Weixian Lei, Qianli Xu, Joo Lim, Mike Shou

Visual Data in NLP

Challenge:	Existing studies on the use of exocentric and egocentric videos in video question answering are focusing on eye-gaze information.
Approach:	They propose a task-oriented VQA dataset that captures eye-gaze information . they propose assisting models that ground the perceptual input into semantic information based on three different answer types .
Outcome:	The proposed model can ground the perceptual input into semantic information while reducing ambiguities.

Beyond Profile: From Surface-Level Facts to Deep Persona Simulation in LLMs (2025.findings-acl)

Copied to clipboard

Zixiao Wang, Duzhen Zhang, Ishita Agarwal, Shen Gao, Le Song, Xiuying Chen

Natural Language Interfaces

Challenge:	Existing approaches to persona simulation large language models (LLMs) focus on learning basic biographical information, or using limited role-play dialogue datasets to capture a character’s responses.
Approach:	They propose to train characters using a linguistic structure and a style-tuning mechanism that allows a general linguistic style expert to collaborate with other task-specific experts to better understand their thoughts.
Outcome:	The proposed model outperforms baselines on linguistic accuracy and opinion comprehension on three tasks for Lu Xun's essay collection.

Multilingual Knowledge Graph Completion via Efficient Multilingual Knowledge Sharing (2025.findings-emnlp)

Copied to clipboard

Cunli Mao, Xiaofei Gao, Ran Song, Shizhu He, Shengxiang Gao, Kang Liu, Zhengtao Yu

Multimodal Machine Translation with Text-Image In-depth Questioning (2025.findings-acl)

Challenge:	Existing MKGC research ignores the shareability of cross-lingual knowledge.
Approach:	They propose a multilingual knowledge Graph Completion framework that leverages multilingual shared knowledge to significantly enhance performance through two components: Knowledge-level Grouped Mixture of Experts (KL-GMoE) and Iterative Entity Reranking (IER).
Outcome:	The proposed framework achieves improvements of 5.47%, 3.27%, and 1.01% in the Hits@1, Hits @3, and Hits_10 metrics, respectively, compared with existing state-of-the-art (SOTA) MKGC method.

Copied to clipboard

Yue Gao, Jing Zhao, Shiliang Sun, Xiaosong Qiao, Tengfei Song, Hao Yang