Papers by Zhiyuan Yu

Program Transfer for Answering Complex Questions over Knowledge Bases (2022.acl-long)

Challenge:	Existing text-guided image editing methods rely on end-to-end pixel-level inpainting paradigm . existing models lack such intermediate representations and Reasoning-then-action process .
Approach:	They propose a "Decompose-then-Action" paradigm that revisits image editing as an actionable interaction process within a structured environment.
Outcome:	The proposed paradigm outperforms existing methods in compositional editing tasks.

Copied to clipboard

Shulin Cao, Jiaxin Shi, Zijun Yao, Xin Lv, Jifan Yu, Lei Hou, Juanzi Li, Zhiyuan Liu, Jinghui Xiao

Knowledge Representation

Challenge:	Program induction for complex questions over knowledge bases relies on a large number of parallel question-program pairs for the given KB, but the gold program annotations are usually lacking, making learning difficult.
Approach:	They propose an approach to leverage program annotations on rich KBs as external supervision signals to aid program induction for low-resourced KB.
Outcome:	The proposed approach outperforms SOTA methods on ComplexWebQuestions and WebQuestionSP.

LLM×MapReduce-V3: Enabling Interactive In-Depth Survey Generation through a MCP-Driven Hierarchically Modular Agent System (2025.emnlp-demos)

Copied to clipboard

Yu Chao, Siyu Lin, Xiaorong Wang, Zhu Zhang, Zihan Zhou, Haoyu Wang, Shuo Wang, Jie Zhou, Zhiyuan Liu, Maosong Sun

Challenge:	Generating high-quality long-form survey articles poses significant challenges to AI Agent systems.
Approach:	They propose a hierarchically modular agent system for long-form survey generation . they use atomic models to implement skeleton initialization, digest construction, and skelet refinement . human evaluations demonstrate system surpasses representative baselines .
Outcome:	The proposed system surpasses representative baselines in both content depth and length, highlighting the strength of MCP-based modular planning.

How Much Do Robots Understand Rudeness? Challenges in Human-Robot Interaction (2024.lrec-main)

Copied to clipboard

Michael Andrew Orme, Yanchao Yu, Zhiyuan Tan

Challenge:	This paper examines the pressing need to understand and manage inappropriate language within the evolving human-robot interaction landscape.
Approach:	They propose to use data cleaning methods to identify inappropriate language in real-time interactions and evaluate natural language models for their proficiency in discerning rudeness.
Outcome:	The proposed methods identify and mitigate inappropriate language in real-time interactions and evaluate natural language models for their proficiency in discerning rudeness.

HeteroCache: A Dynamic Retrieval Approach to Heterogeneous KV Cache Compression for Long-Context LLM Inference (2026.acl-long)

Copied to clipboard

Zhiyuan Shi, Qibo Qiu, null Xuefeng, Zhonglin Jiang, Li Yu, Jian Jiang, Xiaofei He, Wenxiao Wang

KBAlign: Efficient Self Adaptation on Specific Textual Knowledge Bases (2025.findings-emnlp)

Challenge:	Existing static compression methods suffer from coarse-grained caching and high I/O overhead.
Approach:	They propose a training-free dynamic compression framework that uses a sparse attention mechanism to categorize attention heads based on stability and similarity.
Outcome:	The proposed framework achieves state-of-the-art performance on long-context benchmarks and accelerates decoding by up to 3 compared to the original model with a 224K context.

Copied to clipboard

Zheni Zeng, Yuxuan Chen, Shi Yu, Ruobing Wang, Yukun Yan, Zhenghao Liu, Shuo Wang, Xu Han, Zhiyuan Liu, Maosong Sun

Judge as A Judge: Improving the Evaluation of Retrieval-Augmented Generation through the Judge-Consistency of Large Language Models (2025.findings-acl)

Challenge:	Existing methods for retrieval-augmented generation (RAG) are limited and fine-tuning incurs prohibitive costs of external signals.
Approach:	They propose a self-supervised framework that enhances RAG systems through efficient model adaptation.
Outcome:	The proposed framework achieves 90% of the performance gain obtained through GPT-4-supervised adaptation while relying entirely on self-annotation of much smaller models.

Copied to clipboard

Shuliang Liu, Xinze Li, Zhenghao Liu, Yukun Yan, Cheng Yang, Zheni Zeng, Zhiyuan Liu, Maosong Sun, Ge Yu

Rethinking Jailbreak Detection of Large Vision Language Models with Representational Contrastive Scoring (2026.acl-long)

Challenge:	Existing evaluation metrics cannot fairly evaluate the outputs of RAG models during training and evaluation.
Approach:	They propose a method which prompts LLMs to generate different judgments based on various combinations of judgment dimensions and utilizes the judge-consistency to evaluate these judgments.
Outcome:	The proposed method generates more accurate evaluations for RAG models across different RAG model and datasets.

Copied to clipboard

Peichun Hua, Hao Li, Shanghao Shi, Zhiyuan Yu, Ning Zhang

Enhancing Long-Chain Reasoning Distillation through Error-Aware Self-Reflection (2026.findings-acl)

Challenge:	Large Vision-Language Models (LVLMs) are vulnerable to a growing array of multimodal jailbreak attacks, necessitating a generalizable defense that is efficient for practical deployment.
Approach:	They propose a framework that uses a lightweight projection to separate benign and malicious inputs in safety-critical layers.
Outcome:	The proposed framework enables a simple yet powerful contrastive score that differentiates true malicious intent from mere distribution shift.

Copied to clipboard

Zhuoyang Wu, Xinze Li, Zhenghao Liu, Yukun Yan, Zhiyuan Liu, Minghe Yu, Cheng Yang, Yu Gu, Ge Yu, Maosong Sun

RA2FD: Distilling Faithfulness into Efficient Dialogue Systems (2024.emnlp-main)

Challenge:	Existing studies treat SLMs as student models and use long-form Chains-of-Thought (CoTs) as supervision signals for Supervised Fine-Tuning (SFT). Existing research focuses on distilling reasoning ability from LLMs to enhance the mathematical reasoning performance of small-scale models.
Approach:	They propose a framework that refines teacher CoTs through an error-aware reflection process to enable the student model to construct more tailored teacher Cots.
Outcome:	Experiments on multiple mathematical reasoning benchmarks show that ORION improves performance by more than 2% over all baselines.

Copied to clipboard

Zhiyuan Zhu, Yusheng Liao, Chenxin Xu, Yunfeng Guan, Yanfeng Wang, Yu Wang

Green & Sustainable NLP

Challenge:	Retrieval Augmented Generation (RAG) is effective but inference inefficient, while Retrieral Free Generations (RFG) are more efficient but sacrifice faithfulness.
Approach:	They propose a retrieval-free model training scheme that uses a teacher-student framework to distill the faithfulness capacity of a student's knowledge-infused responses.
Outcome:	The proposed model surpasses the previous SOTA RFG model on knowledge-grounded dialogue datasets by an average of 33% while improving inference efficiency.

ExpanRL: Hierarchical Reinforcement Learning for Course Concept Expansion in MOOCs (2020.aacl-main)

Copied to clipboard

Jifan Yu, Chenyu Wang, Gan Luo, Lei Hou, Juanzi Li, Jie Tang, Minlie Huang, Zhiyuan Liu

Challenge:	Existing methods for concept expansion in MOOCs are inefficient because of the diversity of MOOC courses and rapid updates.
Approach:	They propose an end-to-end hierarchical reinforcement learning (HRL) model for concept expansion in MOOCs that employs a two-level mechanism of seed selection and concept expansion.
Outcome:	The proposed model improves on nine real MOOC datasets and maintains competitive performance under different settings.

Reduce Catastrophic Forgetting of Dense Retrieval Training with Teleportation Negatives (2022.emnlp-main)

Copied to clipboard

Si Sun, Chenyan Xiong, Yue Yu, Arnold Overwijk, Zhiyuan Liu, Jie Bao

CE-VDG: Counterfactual Entropy-based Bias Reduction for Video-grounded Dialogue Generation (2024.lrec-main)

Challenge:	Recent research shows strong empirical advantages of dense retrieval in various information access scenarios, such as openQA.
Approach:	They propose a method which accumulates momentum negatives from past iterations and approximates future iteration with lookahead negatives as "teleportations" on web search and OpenQA, ANCE-Tele outperforms previous state-of-the-art systems of similar size and eliminates the dependency on sparse retrieval negatives.
Outcome:	The proposed method outperforms previous state-of-the-art systems on web search and OpenQA and is competitive among systems with significantly more parameters.

Copied to clipboard

Hongcheng Liu, Pingjie Wang, Zhiyuan Zhu, Yanfeng Wang, Yu Wang

Dialogue Systems & Conversational Agents

Challenge:	Existing methods to reduce question-related bias in video-grounded dialogue generation (VDG) however, the dataset often contains inherent bias, which can cause VDG models to learn spurious correlations between questions and answers.
Approach:	They propose to extend the counterfactual reasoning from the information entropy perspective to the generative task, which can effectively reduce the question-related bias in the auto-regressive generation task.
Outcome:	The proposed method can reduce question-related bias in the auto-regressive generation task by using counterfactual entropy as an external loss.

FewRel: A Large-Scale Supervised Few-Shot Relation Classification Dataset with State-of-the-Art Evaluation (D18-1)

Copied to clipboard

Xu Han, Hao Zhu, Pengfei Yu, Ziyun Wang, Yuan Yao, Zhiyuan Liu, Maosong Sun

Low-Resource NLP

Challenge:	Empirical results show that even the most competitive few-shot learning models struggle on this task, especially as compared with humans.
Approach:	They propose a Few-Shot Relation Classification Dataset consisting of 70, 000 sentences on 100 relations derived from Wikipedia and annotated by crowdworkers.
Outcome:	The proposed methods perform well on the most competitive few-shot learning models, especially as compared with humans.

DeepNote: Note-Centric Deep Retrieval-Augmented Generation (2025.findings-emnlp)

Copied to clipboard

Ruobing Wang, Qingfei Zhao, Yukun Yan, Daren Zha, Yuxuan Chen, Shi Yu, Zhenghao Liu, Yixuan Wang, Shuo Wang, Xu Han, Zhiyuan Liu, Maosong Sun

MatPlotAgent: Method and Evaluation for LLM-Based Agentic Scientific Data Visualization (2024.findings-acl)

Challenge:	. - (EN)
Approach:	. - (EN)
Outcome:	. - (EN)

Copied to clipboard

Zhiyu Yang, Zihan Zhou, Shuo Wang, Xin Cong, Xu Han, Yukun Yan, Zhenghao Liu, Zhixing Tan, Pengyuan Liu, Dong Yu, Zhiyuan Liu, Xiaodong Shi, Maosong Sun

SimPBL: A Multi-Agent Framework for Project-Based Learning (2026.acl-long)

Challenge:	Scientific data visualization is an essential process in research, but its use of large language models remains unexplored.
Approach:	They propose a model-agnostic LLM agent framework to automate scientific data visualization tasks.
Outcome:	The proposed framework improves performance of commercial and open-source models.

Copied to clipboard

Daniel Zhang-Li, Joy Jia Yin Lim, Binglin Liu, Shangqing Tu, Zijun Yao, Hao Peng, Jifan Yu, Haoxuan Li, Zhanxin Hao, Ye He, Zekun Li, Jiangyi Wang, Lei Hou, Bin Xu, Xin Cong, Zhiyuan Liu, Huiqin Liu, Yu Zhang, Juanzi Li

Challenge:	Existing LLMs provide partial assistance without modeling these roles, and overly comprehensive help can reduce learner autonomy.
Approach:	They propose a multi-agent framework with an orchestrator agent that provides adaptive scaffolding from interaction logs and collaborator agents that support project work through boundary-aware collaboration.
Outcome:	The proposed framework improves learner examination scores by 14% . it is based on a multi-agent framework with an orchestrator agent .

Comprehensive Benchmarking of Long-Form Speech Generation in Diverse Scenarios (2026.findings-acl)

Copied to clipboard

Changhao Pan, Rui Yang, Han Wang, Zhuan Zhou, Xuming He, Wenxiang Guo, Ziyue Jiang, Ruiqi Li, Yu Zhang, Chenyuhao Wen, Ke Lei, Xiang Yin, Jingyu Lu, Zhiyuan Zhu, Zhou Zhao

Structure-Aware Language Model Pretraining Improves Dense Retrieval on Structured Data (2023.findings-acl)

Challenge:	Existing evaluation benchmarks for long-form speech are limited to limited domains, creating a significant gap with the diverse downstream applications.
Approach:	They propose a benchmark that decomposes "long-form speech quality" into specific, disentangled dimensions.
Outcome:	The proposed benchmark decomposes “long-form speech quality” into specific, disentangled dimensions.

Copied to clipboard

Xinze Li, Zhenghao Liu, Chenyan Xiong, Shi Yu, Yu Gu, Zhiyuan Liu, Ge Yu

INTERVENOR: Prompting the Coding Ability of Large Language Models with the Interactive Chain of Repair (2024.findings-acl)

Challenge:	Structure Aware Dense Retrieval (SANTA) model encodes user queries and structured data in one universal embedding space for retrieving structured data.
Approach:	They propose to use structured data and unstructured data to encode queries and structured data in one universal embedding space for retrieving structured data.
Outcome:	The proposed model achieves state-of-the-art on code search and product search and conducts convincing results in the zero-shot setting.

Copied to clipboard

Hanbin Wang, Zhenghao Liu, Shuo Wang, Ganqu Cui, Ning Ding, Zhiyuan Liu, Ge Yu

MOOCCube: A Large-scale Data Repository for NLP Applications in MOOCs (2020.acl-main)

Challenge:	Experimental results show that INTERVENOR surpasses baseline models, exhibiting improvements of approximately 18% and 4.3% over GPT-3.5 in code generation and code translation tasks.
Approach:	They propose a system that prompts Large Language Models to play distinct roles during the code repair process, functioning as both a Code Learner and a code teacher.
Outcome:	The proposed system surpasses baseline models in code generation and code translation tasks and improves on syntax errors and assertion errors.

Copied to clipboard

Jifan Yu, Gan Luo, Tong Xiao, Qingyang Zhong, Yuquan Wang, Wenzheng Feng, Junyi Luo, Chenyu Wang, Lei Hou, Juanzi Li, Zhiyuan Liu, Jie Tang

Craw4LLM: Efficient Web Crawling for LLM Pretraining (2025.findings-acl)

Challenge:	Massive open online courses (MOOCs) are a popular educational platform for advanced research.
Approach:	They propose to use MOOCCube to build a large-scale data repository of over 700 MOOC courses, 100k concepts, 8 million student behaviors with an external resource.
Outcome:	The proposed datasets show that they can facilitate research in MOOCs.

Copied to clipboard

Shi Yu, Zhiyuan Liu, Chenyan Xiong

Green & Sustainable NLP

Challenge:	Existing work discards over 90% of the raw data collected from web crawls, highlighting the inefficiency of current web crawlers in collecting LLM pretraining data.
Approach:	They propose a web crawling method that leverages the preference of LLMs as the priority score of the web crawler’s scheduler to obtain high-quality pretraining data.
Outcome:	The proposed method achieves high-quality pretraining data on a web graph containing 900 million webpages from a commercial search engine's index with just 21% URLs crawled.

ThinkNote: Enhancing Knowledge Integration and Utilization of Large Language Models via Constructivist Cognition Modeling (2026.findings-eacl)

Copied to clipboard

Zhipeng Xu, Zhenghao Liu, Yukun Yan, Shuo Wang, Shi Yu, Zheni Zeng, Chaojun Xiao, Zhiyuan Liu, Ge Yu, Chenyan Xiong

A Query-Response Framework for Whole-Page Complex-Layout Document Image Translation with Relevant Regional Concentration (2025.findings-acl)

Challenge:	Large Language Models (LLMs) exhibit suboptimal behaviors and inconsistencies when exposed to unfamiliar external information, underscoring their limitations in effectively leveraging such knowledge.
Approach:	They propose a framework that enhances the external knowledge utilization of Large Language Models through a two-stage constructivist cognitive modeling process.
Outcome:	The proposed framework achieves a 10% improvement over baseline methods on various question-answering benchmarks.

Copied to clipboard

Zhiyang Zhang, Yaping Zhang, Yupu Liang, Zhiyuan Chen, Lu Xiang, Yang Zhao, Yu Zhou, Chengqing Zong

Machine Translation

Challenge:	Existing methods for document image translation rely on the vanilla encoder-decoder paradigm . a novel dynamic aggregation mechanism is designed to enhance the text semantics in query features toward translation.
Approach:	They propose a Query-Response DIT framework that reformulates the DIT task into a parallel response/translation process of multiple queries.
Outcome:	The proposed framework improves translation quality on four translation directions on three benchmarks.

AgentRM: Enhancing Agent Generalization with Reward Modeling (2025.acl-long)

Copied to clipboard

Yu Xia, Jingru Fan, Weize Chen, Siyu Yan, Xin Cong, Zhong Zhang, Yaxi Lu, Yankai Lin, Zhiyuan Liu, Maosong Sun

Robustness in NLP

Challenge:	Existing LLM-based agents have strong performance on held-in tasks, but their generalizability to unseen tasks remains poor.
Approach:	They propose a reward-based generalizable reward model to guide the policy model for effective test-time search.
Outcome:	The proposed agentRM outperforms existing agents on held-in tasks by 8.8 points on average.

Cross-Modal Coreference Alignment: Enabling Reliable Information Transfer in Omni-LLMs (2026.acl-long)

Copied to clipboard

Hongcheng Liu, Yuhao Wang, Zhe Chen, Pingjie Wang, Zhiyuan Zhu, Yixuan Hou, Yanfeng Wang, Yu Wang

LoRE: Enhancing Search Relevance with Progressive Chain-of-Thought and Preference Alignment (2026.findings-acl)

Challenge:	Experiments on 13 Omni-LLMs reveal systematic weaknesses in cross-modal coreference . cross-module coreference is a crucial missing piece for advancing robust omni-modal reasoning.
Approach:	They propose a cross-modal coreference problem to evaluate and enhance Omni-LLMs' reasoning capabilities.
Outcome:	Experiments on 13 Omni-LLMs show they lack coreference-aware thinking patterns . the CROSSOMNI dataset yields significant performance gains and generalizes well to collaborative reasoning tasks.

Copied to clipboard

Chenji Lu, Zhuo Chen, Hui Zhao, Zhiyuan Zeng, Gang Zhao, Junjie Ren, null Lihaoran, Songyan Liu, Pengjie Wang, Chuan Yu, Jian Xu, Bo Zheng

Fusion-in-T5: Unifying Variant Signals for Simple and Effective Document Ranking with Attention Fusion (2024.lrec-main)

Challenge:	E-commerce search relevance is a critical component of retrieval systems.
Approach:	They propose a large-generative model for search relevance that trains reasoning knowledge, multi-modal understanding and rule awareness into three core competencies.
Outcome:	The proposed model outperforms GPT-5 in Macro-F1 and achieves 27% online gain.

Copied to clipboard

Shi Yu, Chenghao Fan, Chenyan Xiong, David Jin, Zhiyuan Liu, Zhenghao Liu

DIAG-NRE: A Neural Pattern Diagnosis Framework for Distantly Supervised Neural Relation Extraction (P19-1)

Challenge:	Current document ranking pipelines involve multiple ranking layers to integrate different information step-by-step.
Approach:	They propose a novel re-ranker Fusion-in-T5 which integrates text matching information, ranking features, and global document information into one single unified model via templated-based input and global attention.
Outcome:	The proposed model significantly improves ranking performance over complex cascade pipelines.

Copied to clipboard

Shun Zheng, Xu Han, Yankai Lin, Peilin Yu, Lu Chen, Ling Huang, Zhiyuan Liu, Wei Xu

Utility-Oriented Visual Evidence Selection for Multimodal Retrieval-Augmented Generation (2026.acl-long)

Challenge:	Existing methods for labeling relational facts require significant expert labor to write relation-specific patterns, which makes them too sophisticated to generalize quickly.
Approach:	They propose a neural pattern diagnosis framework that can summarize and refine relation-specific patterns with human experts in the loop.
Outcome:	The proposed framework can summarize and refine high-quality relational patterns from noise data with human experts in the loop.

Copied to clipboard

Weiqing Luo, Zongye Hu, Xiao Wang, Zhiyuan Yu, Haofeng Zhang, Ziyi Huang

ARISE: An Adaptive Resolution-Aware Metric for Test-Time Scaling Evaluation in Large Reasoning Models (2026.findings-acl)

Challenge:	Existing methods for multimodal retrieval-augmented generation rely on semantic relevance or surface-level similarity, which are often misaligned with the actual utility of visual evidence for downstream reasoning.
Approach:	They propose a latent notion of evidence usefulness and propose 'surrogate-accelerated' framework that efficiently estimates evidence utility using lightweight multimodal models.
Outcome:	The proposed framework outperforms state-of-the-art models while achieving substantial reductions in computational cost.

Copied to clipboard

Zhangyue Yin, Qiushi Sun, Zhiyuan Zeng, Zhiyuan Yu, Qipeng Guo, Xuanjing Huang, Xipeng Qiu

EvolveBench: A Comprehensive Benchmark for Assessing Temporal Awareness in LLMs on Evolving Knowledge (2025.acl-long)

Challenge:	Existing evaluation methods for test-time scaling are limited.
Approach:	They propose an adaptive resolution-aware scaling evaluation metric specifically designed to assess the test-time scaling effectiveness of large reasoning models.
Outcome:	The proposed metric provides a reliable and fine-grained measurement of test-time scaling capabilities, revealing significant variations in scaling efficiency across models.

Copied to clipboard

Zhiyuan Zhu, Yusheng Liao, Zhe Chen, Yuhao Wang, Yunfeng Guan, Yanfeng Wang, Yu Wang

Challenge:	Existing studies have explored how LLMs perceive time, but they often overlook the critical aspect of knowledge utilization.
Approach:	They propose a benchmark that evaluates temporal competence along five key dimensions: Cognition, Awareness, Trustworthiness and reasoning.
Outcome:	EvolveBench measures temporal competence along five key dimensions: Cognition, Awareness, Trustworthiness, Understanding and reasoning.

ReaGeo: Reasoning-Enhanced End-to-End Geocoding with LLMs (2026.findings-acl)

Copied to clipboard

Jian Cui, Zhiyuan Ren, Desheng Weng, Yongqi Zhao, Gong Wenbin, Yu Lei, Zhenning Dong

ReflectRM: Boosting Generative Reward Models via Self-Reflection within a Unified Judgment Framework (2026.acl-long)

Challenge:	Existing methods rely on text retrieval and geographic knowledge bases to generate coordinates, and they are prone to error propagation and dependency on structured knowledge bases.
Approach:	They propose to use large language models to convert geographic coordinates into geohash sequences and introduce a Chain-of-Thought mechanism to enhance the model’s reasoning over spatial relationships.
Outcome:	The proposed framework can handle explicit address queries in single-point predictions and effectively resolve vague relative location queries.

Copied to clipboard

Kai Qin, Liangxin Liu, Yu Liang, Longzheng Wang, null Wangyan, Zhang Yueyang, Long Xia, Zhiyuan Sun, Houde Liu, Daiting Shi

Challenge:	Existing methods for generating reward models focus on outcome-level supervision, neglecting analytical process quality, which constrains their potential.
Approach:	They propose a novel reward model that leverages self-reflection to assess analytical quality and enhance preference modeling.
Outcome:	The proposed model improves performance on four benchmarks and significantly mitigates positional bias.

Integrating Representation Subspace Mapping with Unimodal Auxiliary Loss for Attention-based Multimodal Emotion Recognition (2024.lrec-main)

Copied to clipboard

Xulong Du, Xingnan Zhang, Dandan Wang, Yingying Xu, Zhiyuan Wu, Shiqing Zhang, Xiaoming Zhao, Jun Yu, Liangliang Lou

Sentiment Analysis

Challenge:	Existing methods to identify emotions rely on a large modality gap in their representations .
Approach:	They propose a representation subspace mapping module that maps each modality into two distinct subspaces and a cross-modality attention module that leverages auxiliary loss to remove the noise unrelated to emotion classification.
Outcome:	The proposed approach achieves superior performance to state-of-the-art MER methods on the IEMOCAP and MSP-Improv datasets.

Augmentation-Adapted Retriever Improves Generalization of Language Models as Generic Plug-In (2023.acl-long)

Copied to clipboard

Zichun Yu, Chenyan Xiong, Shi Yu, Zhiyuan Liu

Versatile Framework for Song Generation with Prompt-based Control (2025.findings-emnlp)

Challenge:	Prior work on retrieval augmentation fine-tuned the retriever and the LM, making them closely coupled.
Approach:	They propose a generic retrieval plug-in that can be used to fine-tune retrieval augmentation and a LM to learn a user's preferences.
Outcome:	The proposed retriever improves the generalization of large language models on the MMLU and PopQA datasets by learning LM’s preferences from a known source LM .

Copied to clipboard

Yu Zhang, Wenxiang Guo, Changhao Pan, Zhiyuan Zhu, Ruiqi Li, Jingyu Lu, Rongjie Huang, Ruiyuan Zhang, Zhiqing Hong, Ziyue Jiang, Zhou Zhao

ConsistRM: Improving Generative Reward Models via Consistency-Aware Self-Training (2026.acl-long)

Challenge:	Existing methods for song generation fail to generate vocals with prompt-based control and proper alignment.
Approach:	VersBand is a multi-task song generation framework for synthesizing high-quality songs with prompt-based control.
Outcome:	Experimental results show that VersBand performs better than baseline models across multiple song generation tasks.

Copied to clipboard

Yu Liang, Liangxin Liu, Longzheng Wang, null Wangyan, Zhang Yueyang, Long Xia, Zhiyuan Sun, Daiting Shi

Challenge:	ConsistRM is a self-training framework that enables effective and stable GRM training without human annotations.
Approach:	They propose a self-training framework that enables effective and stable GRM training without human annotations.
Outcome:	The proposed framework outperforms vanilla Reinforcement Fine-Tuning (RFT) by 1.5% on five benchmark datasets.

TCSinger 2: Customizable Multilingual Zero-shot Singing Voice Synthesis (2025.findings-acl)

Copied to clipboard

Yu Zhang, Wenxiang Guo, Changhao Pan, Dongyu Yao, Zhiyuan Zhu, Ziyue Jiang, Yuhan Wang, Tao Jin, Zhou Zhao

Responsible & Trustworthy NLP

Challenge:	Existing zero-shot singing voice synthesis models depend on phoneme and note boundary annotations, limiting their robustness and producing poor transitions between phonemes and notes.
Approach:	They propose a multi-task multilingual zero-shot SVS model with style transfer and style control based on various prompts.
Outcome:	Experimental results show that TCSinger 2 outperforms baseline models in subjective and objective metrics across multiple related tasks.

Dimension Reduction for Efficient Dense Retrieval via Conditional Autoencoder (2022.emnlp-main)

Copied to clipboard

Zhenghao Liu, Han Zhang, Chenyan Xiong, Zhiyuan Liu, Yu Gu, Xiaohua Li

Thermometer of Thoughts: Enhancing LLM’s Exploration via Attention Temperature Modulation (2026.acl-long)

Challenge:	Existing work reserves the principle dimensions of query and document embeddings for building more efficient retrieval systems.
Approach:	They propose to use Conditional Autoencoder to compress high-dimensional embeddings to maintain the same embeddable distribution and better recover ranking features.
Outcome:	The proposed algorithm achieves comparable ranking performance with its teacher model and makes the retrieval system more efficient.

Copied to clipboard

Zhiyuan Yu, Shijian Xiao, Cam-Tu Nguyen, Zhangyue Yin, Lekai Xing, Wenzhong Li, Sanglu Lu

Simulating Classroom Education with LLM-Empowered Agents (2025.naacl-long)

Challenge:	Recent advances in the reasoning capabilities of large language models have enabled them to tackle complex tasks such as mathematics reasoning.
Approach:	They propose a method that modulates attention temperature dynamically to shift LLM's internal focus during reasoning, enabling a dynamic shift between exploratory and focused modes.
Outcome:	The proposed method improves Pass@10 by 6.78–14.20% and aggregation accuracy by 9.74% across 7 reasoning benchmarks.

Copied to clipboard

Zheyuan Zhang, Daniel Zhang-Li, Jifan Yu, Linlu Gong, Jinchang Zhou, Zhanxin Hao, Jianxiao Jiang, Jie Cao, Huiqin Liu, Zhiyuan Liu, Lei Hou, Juanzi Li

TSPO: Breaking the Double Homogenization Dilemma in Multi-turn Search Policy Optimization (2026.findings-acl)

Challenge:	Initial studies have focused on task-specific, independent LLM-empowered agents, but the potential of LLMs within a multi-agent collaborative framework for classroom simulation with real user participation remains unexplored.
Approach:	They propose a multi-agent classroom simulation teaching framework that recognizes representative class roles and introduces a novel class control mechanism for automatic classroom teaching.
Outcome:	The proposed framework can simulate dynamic learning environment for users with active teacher-student and student-studente interactions.

Copied to clipboard

Shichao Ma, Zhiyuan Ma, Ming Yang, Xiaofan Li, Xing Wu, Jintao Du, Yu Cheng, Weiqiang Wang, Qiliang Liu, Zhengyang Zhou, Yang Wang

HeteroRAG: A Heterogeneous Retrieval-Augmented Generation Framework for Medical Vision Language Tasks (2026.findings-acl)

Challenge:	Large Language Models (LLMs) can solve complex tasks through iterative information retrieval.
Approach:	They propose a turn-level stage-aware policy optimization approach to solve this problem . they introduce a first-occurrence latent reward mechanism to allocate partial rewards .
Outcome:	Experiments show that TSPO outperforms state-of-the-art models on Qwen2.5-3B and 7B models.

Copied to clipboard

Zhe Chen, Yusheng Liao, Zhiyuan Zhu, Haolin Li, Hongcheng Liu, Yanfeng Wang, Yu Wang

Improving MLLM’s Document Image Machine Translation via Synchronously Self-reviewing Its OCR Proficiency (2025.findings-acl)

Challenge:	Medical large vision-language models suffer from factual inaccuracies and unreliable outputs.
Approach:	They propose a framework that enhances Med-LVLMs through heterogeneous knowledge sources.
Outcome:	The proposed framework improves Med-LVLMs through heterogeneous knowledge sources.

Copied to clipboard

Yupu Liang, Yaping Zhang, Zhiyang Zhang, Zhiyuan Chen, Yang Zhao, Lu Xiang, Chengqing Zong, Yu Zhou

Multilinguality

Challenge:	Multimodal Large Language Models (MLLMs) have shown strong performance in document image tasks, especially Optical Character Recognition (OCR). However, they struggle with Document Image Machine Translation (DIMT), which requires handling both cross-modal and cross-lingual challenges.
Approach:	They propose a novel fine-tuning paradigm that allows the model to generate OCR text before producing translation text, which allows it to leverage its strong monolingual OCR ability while learning to translate text across languages.
Outcome:	The proposed model can leverage its strong monolingual OCR ability while learning to translate text across languages.

Hierarchical Relation Extraction with Coarse-to-Fine Grained Attention (D18-1)

Copied to clipboard

Xu Han, Pengfei Yu, Zhiyuan Liu, Maosong Sun, Peng Li

LM-Interview: An Easy-to-use Smart Interviewer System via Knowledge-guided Language Model Exploitation (2024.emnlp-demo)

Challenge:	Existing methods for relation extraction use knowledge graphs to automatically label training data . but, it suffers from the wrong labeling problem because not all sentences containing two entities can express their relations in KGs .
Approach:	They propose a distant supervision approach to automatically label training instances . they integrate hierarchical information of relations into distantly supervised relation extraction .
Outcome:	The proposed model outperforms baseline models on a large-scale dataset.

Copied to clipboard

Hanming Li, Jifan Yu, Ruimiao Li, Zhanxin Hao, Yan Xuan, Jiaxi Yuan, Bin Xu, Juanzi Li, Zhiyuan Liu

Multi-Modal Multi-Granularity Tokenizer for Chu Bamboo Slips (2025.coling-main)

Challenge:	Semi-structured interviews are a crucial method of data acquisition in qualitative research.
Approach:	They propose a semi-structured interview system that automates interview preparation, analysis and control by interviewers.
Outcome:	Experimental results show that LM-Interview performs comparable to human interviewers . the system can be used to analyze semi-structured interviews without interviewers' involvement .

Copied to clipboard

Yingfa Chen, Chenlong Hu, Cong Feng, Chenyang Song, Shi Yu, Xu Han, Zhiyuan Liu, Maosong Sun

Syntactic Text Processing

Challenge:	Using a multi-modal multi-granularity tokenizer, we analyze ancient Chinese scripts . a large proportion of the characters in ancient Chinese are rare or undeciphered .
Approach:	They propose a multi-modal multi-granularity tokenizer specifically designed for ancient Chinese scripts.
Outcome:	The proposed tokenizer improves on the part-of-speech tagging task on the Chu bamboo slip script.

COAST: Enhancing the Code Debugging Ability of LLMs through Communicative Agent Based Data Synthesis (2025.findings-naacl)

Copied to clipboard

Weiqing Yang, Hanbin Wang, Zhenghao Liu, Xinze Li, Yukun Yan, Shuo Wang, Yu Gu, Minghe Yu, Zhiyuan Liu, Ge Yu

PRESTO: Progressive Pretraining Enhances Synthetic Chemistry Outcomes (2024.findings-emnlp)

Challenge:	Existing code debugging benchmarks focus on the Code Repair stage of the code generation process.
Approach:	They propose a framework to evaluate the debugging abilities of large language models by emulating the human debug process.
Outcome:	The proposed framework outperforms human-curated and GPT-4-generated training data, enabling 7B-scale LLMs to achieve comparable debugging performance to GPT-3.5.

Copied to clipboard

He Cao, Yanjun Shao, Zhiyuan Liu, Zijing Liu, Xiangru Tang, Yuan Yao, Yu Li

MAXS: Meta-Adaptive Exploration with LLM Agents (2026.findings-acl)

Challenge:	Multimodal Large Language Models (MLLMs) have seen growing adoption across various scientific domains.
Approach:	They propose a framework that bridges the molecule-text modality gap by integrating a comprehensive benchmark of pretraining strategies and dataset configurations.
Outcome:	The proposed framework improves multimodal LLMs through cross-modal alignment and multi-graph understanding.

Copied to clipboard

Jian Zhang, Zhiyuan Wang, Zhangqi Wang, Yu He, Haoran Luo, li Yuan, Lingling Zhang, Rui Mao, Qika Lin, Jun Liu

Responsible & Trustworthy NLP

Challenge:	Existing methods for inference are often myopic and have divergent reasoning paths . a meta-adaptive reasoning framework is proposed to improve the efficiency of LLM agents .
Approach:	They propose a meta-adaptive reasoning framework that integrates tool execution and reasoning planning.
Outcome:	The proposed framework outperforms existing methods in performance and inference efficiency.

Course Concept Expansion in MOOCs with External Knowledge and Interactive Game (P19-1)

Copied to clipboard

Jifan Yu, Chenyu Wang, Gan Luo, Lei Hou, Juanzi Li, Zhiyuan Liu, Jie Tang

Knowledge Representation

Challenge:	Existing methods to expand course concepts in MOOCs suffer from semantic drifts and lack of knowledge guidance.
Approach:	They propose to use a boundary search method to search for new concepts via external knowledge base and then use heterogeneous features to verify the results.
Outcome:	The proposed method improves on the datasets from Coursera and XuetangX.

LLM×MapReduce: Simplified Long-Sequence Processing using Large Language Models (2025.acl-long)

Copied to clipboard

Zihan Zhou, Chong Li, Xinyi Chen, Shuo Wang, Yu Chao, Zhili Li, Haoyu Wang, Qi Shi, Zhixing Tan, Xu Han, Xiaodong Shi, Zhiyuan Liu, Maosong Sun

Unleashing Low-Bit Inference on Ascend NPUs: A Comprehensive Evaluation of HiFloat Formats (2026.acl-industry)

Challenge:	Existing studies have focused on extending the context length of large language models (LLMs) due to their quadratic computational complexity and a lack of high-quality long training examples, most LLMs are trained with a limited window size.
Approach:	They propose a training-free framework that enables large language models to effectively process long texts using a divide-and-conquer strategy for comprehensive document understanding.
Outcome:	The proposed framework outperforms open-source and commercial long-context LLMs and is compatible with several models.

Copied to clipboard

Pengxiang Zhao, Hui-Ling Zhen, Xing Li, Han Bao, Weizhe Lin, Zhiyuan Yang, Yu Zi Wei, Xin Wang, Mingxuan Yuan, Xianzhi Yu, Zhenhua Dong

Green & Sustainable NLP

Challenge:	Low-bit floating-point formats like MXFP and NVFP4 offer new opportunities for precision and efficiency.
Approach:	They evaluate HiFloat (HiF8 and HiF4), a family of floating-point formats tailored for Ascend NPUs.
Outcome:	The proposed formats excel with high-variance data and are compatible with state-of-the-art quantization frameworks.

Enhancing Open-Domain Task-Solving Capability of LLMs via Autonomous Tool Integration from GitHub (2025.acl-long)

Copied to clipboard

Bohan Lyu, Xin Cong, Heyang Yu, Pan Yang, Cheng Qian, Zihe Wang, Yujia Qin, Yining Ye, Yaxi Lu, Chen Qian, Zhong Zhang, Yukun Yan, Yankai Lin, Zhiyuan Liu, Maosong Sun