Papers by Xinyu Yin
Reinforced Efficient Reasoning via Semantically Diverse Exploration (2026.acl-long)
Copied to clipboard
Ziqi Zhao, Zhaochun Ren, Jiahong Zou, Liu Yang, Zhiwei Xu, Xuri Ge, Zhumin Chen, Xinyu Ma, Daiting Shi, Shuaiqiang Wang, Dawei Yin, Xin Xin
| Challenge: | Existing methods for reinforcement learning with verifiable rewards suffer from limited exploration diversity and inefficient reasoning. |
| Approach: | They propose a method that rewards concise and correct reasoning while penalizing unnecessarily long reasoning chains. |
| Outcome: | Extensive experiments on Qwen and Llama models validate the effectiveness and efficiency of ROSE. |
CoRanking: Collaborative Ranking with Small and Large Ranking Agents (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Listwise ranking based on Large Language Models (LLMs) has achieved state-of-the-art performance in Information Retrieval (IR) however, their effectiveness often depends on LLMs with massive parameter scales and computationally expensive sliding window processing, leading to substantial efficiency bottlenecks. |
| Approach: | They propose a Collaborative Ranking framework (CoRanking) for LLM-based listwise ranking based on large language models with massive parameter scales and computationally expensive sliding window processing. |
| Outcome: | The proposed framework reduces ranking latency by approximately 70% while improving effectiveness compared to the standalone large reranker. |
Aligning Large Language Models via Fully Self-Synthetic Data (2026.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to reinforcement learning from human feedback (RLHF) require expensive human-annotated datasets and proprietary models like GPT-4 to annotate preference pairs. |
| Approach: | They propose a self-synthetic framework for LLM alignment where all training data, including prompts (i.e., user queries), responses, and preferences, are generated by the model itself. |
| Outcome: | The proposed framework enhances the model’s chat capabilities on standard benchmarks like AlpacaEval 2.0 while maintaining strong performance on downstream objective tasks. |
Themis: A Reference-free NLG Evaluation Language Model with Flexibility and Interpretability (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for evaluation of natural language generation tasks lack reliable data. |
| Approach: | They propose to use annotations from human and GPT-4 to construct a corpus for NLG evaluation. |
| Outcome: | The proposed corpus can perform flexible and interpretable evaluations without references and surpasses existing models. |
Exploring Context-Aware Evaluation Metrics for Machine Translation (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing studies on machine translation evaluation focused on quality of individual sentences, while neglecting the importance of contextual information. |
| Approach: | They propose a context-aware machine translation evaluation metric called Cont-COMET . they use the COMET framework to consider the preceding and subsequent contexts of the sentence . |
| Outcome: | The proposed metric improves system-level and segment-level evaluations on the official WMT framework. |
The Mirage of Model Editing: Revisiting Evaluation in the Wild (2025.acl-long)
Copied to clipboard
| Challenge: | despite near-perfect results, effectiveness of model editing in real-world applications remains unclear. |
| Approach: | They propose QAEdit and WILD to better reflect real-world use of model editing . they propose a benchmark aligned with widely used question answering datasets and a task-agnostic evaluation framework . |
| Outcome: | The proposed QAEdit benchmark and WILD evaluation framework show that current models perform worse than previously reported. |
Towards General Agentic Intelligence via Environment Scaling (2026.findings-acl)
Copied to clipboard
Runnan Fang, Shihao Cai, Baixuan Li, Jialong Wu, Guangyu Li, Wenbiao Yin, Xinyu Wang, Xiaobin Wang, Liangcai Su, Zhen Zhang, Shibin Wu, Zhengwei Tao, Yong Jiang, Pengjun Xie, Ningyu Zhang, Fei Huang, Wentao Zhang, Jingren Zhou
| Challenge: | Diverse real-world APIs require precise, robust function-calling intelligence, which needs agents to develop these capabilities through interaction in varied environments. |
| Approach: | They propose a framework that scales up environments to enable agentic intelligence . they use a two-phase agent fine-tuning strategy to first endow agents with basic agentic capabilities, then specializing them for domain-specific contexts. |
| Outcome: | Experiments on -bench, -Bench, and ACEBench show that the model significantly enhances the models’ function-calling capability. |
MC-MKE: A Fine-Grained Multimodal Knowledge Editing Benchmark Emphasizing Modality Consistency (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing benchmarks for knowledge editing in multimodal large language models focus on limited scenarios due to the lack of rigorous definition of multimodal knowledge. |
| Approach: | They propose a decomposed definition of multimodal knowledge and a benchmark to evaluate it. |
| Outcome: | The proposed method reveals that it is difficult to define multimodal knowledge editing in LLMs. |
ReasonRank: Empowering Passage Ranking with Strong Reasoning Ability (2026.acl-long)
Copied to clipboard
| Challenge: | Existing rerankers perform poorly in complex ranking scenarios due to the scarcity of reasoning-intensive training data. |
| Approach: | They propose an automated reasoning-intensive training framework which generates high-quality training labels from training queries and passages. |
| Outcome: | The proposed model outperforms baselines significantly and achieves much lower latency than the pointwise reranker. |
AutoConv: Automatically Generating Information-seeking Conversations with Large Language Models (2023.acl-short)
Copied to clipboard
Siheng Li, Cheng Yang, Yichun Yin, Xinyu Zhu, Zesen Cheng, Lifeng Shang, Xin Jiang, Qun Liu, Yujiu Yang
| Challenge: | Existing research on information-seeking conversations is stymied by the lack of training data. |
| Approach: | They propose to use autoconv for synthetic conversation generation to capture the characteristics of the information-seeking process and fine tune an LLM with a few human conversations to generate synthetic conversations with high quality. |
| Outcome: | The proposed model improves on two commonly-used datasets and alleviates the dependence on human annotation. |
Is ChatGPT Good at Search? Investigating Large Language Models as Re-Ranking Agents (2023.emnlp-main)
Copied to clipboard
Weiwei Sun, Lingyong Yan, Xinyu Ma, Shuaiqiang Wang, Pengjie Ren, Zhumin Chen, Dawei Yin, Zhaochun Ren
| Challenge: | Existing work utilizes generative LLMs for Information Retrieval (IR) rather than direct passage ranking. |
| Approach: | They investigate generative LLMs such as ChatGPT and GPT-4 for relevance ranking in IR and use a test set to verify the model’s ability to rank unknown knowledge. |
| Outcome: | The proposed model outperforms a 3B supervised model on the BEIR benchmark. |
TP-RAG: Benchmarking Retrieval-Augmented Large Language Model Agents for Spatiotemporal-Aware Travel Planning (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies on large language models (LLMs) focus on basic plan validity, but neglect critical aspects such as route efficiency, POI appeal, and real-time adaptability. |
| Approach: | They propose a benchmark for retrieval-augmented, spatiotemporal-aware travel planning that integrates retrieved trajectories with LLMs’ intrinsic reasoning. |
| Outcome: | The proposed framework improves spatial efficiency and POI rationality while challenging universality and robustness due to conflicting references and noisy data. |
Evaluating Self-Generated Documents for Enhancing Retrieval-Augmented Generation with Large Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Recent research focuses on optimizing the use of Self-Docs with their inherent properties remaining underexplored. |
| Approach: | They develop a taxonomy to compare the effectiveness of different types of Self-Docs and explore strategies for combining them with external sources. |
| Outcome: | The proposed model can supplement retrieved content and provide a powerful way to improve knowledge-intensive question answering tasks. |
Agentic-R: Learning to Retrieve for Agentic Search (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing retrievers for single-turn retrieval-augmented generation (RAG) rely on similarity-based retrievers, but similar passages are not always useful for final answer generation. |
| Approach: | They propose a retrieval-augmented-generation retriever that integrates reasoning with retrieval . they use local query-passage relevance and global answer correctness to measure passage utility . |
| Outcome: | The proposed retriever outperforms existing retrievers on QA benchmarks on seven single-hop and multi-hop searches. |
Error-Robust Retrieval for Chinese Spelling Check (2024.lrec-main)
Copied to clipboard
| Challenge: | Chinese Spelling Check (CSC) aims to detect and correct spelling errors in Chinese texts . current methods may not fully leverage existing datasets, resulting in insufficient annotated data . |
| Approach: | They propose a plug-and-play retrieval method with error-robust information for Chinese Spelling Check . they employ multimodal representations that fuse phonetic, morphologic, and contextual information . |
| Outcome: | The proposed method improves on the SIGHAN benchmarks on Chinese spelling check (CSC) the proposed method is based on training data and lacks adequate parallel corpora . |
MTR-Suite: A Framework for Evaluating and Synthesizing Conversational Retrieval Benchmarks (2026.acl-long)
Copied to clipboard
Junhao Ruan, Abudukeyumu Abudula, Bei Li, Yongjing Yin, Xinyu Liu, Kechen Jiao, Xin Chen, Jingang Wang, Xunliang Cai, Tong Xiao, JingBo Zhu
| Challenge: | Existing conversational retrieval benchmarks suffer from costly, sparse human annotation or rigid, unnatural automated heuristics. |
| Approach: | They propose a framework for auditing, synthesizing, and benchmarking conversational retrieval. |
| Outcome: | The proposed framework is based on three LLM-based auditors and a multi-agent system . it mimics production-style challenges (hard topic switching, verbosity) and offers superior discriminative power. |
Sliding Windows Are Not the End: Exploring Full Ranking with Long-Context Large Language Models (2025.acl-long)
Copied to clipboard
| Challenge: | Existing methods for listwise passage ranking use sliding window approach, which is inefficient as it requires repetitive and serialized processing. |
| Approach: | They propose a listwise label construction approach and importance-aware learning objective for full ranking. |
| Outcome: | The proposed method outperforms existing methods in listwise ranking tasks. |
TransFace: Unit-Based Audio-Visual Speech Synthesizer for Talking Head Translation (2024.findings-acl)
Copied to clipboard
Xize Cheng, Rongjie Huang, Linjun Li, Zehan Wang, Tao Jin, Aoxiong Yin, Chen Feiyang, Xinyu Duan, Baoxing Huai, Zhou Zhao
| Challenge: | Existing methods for talking head translation rely on cascading, resulting in delays and cascadic errors. |
| Approach: | They propose a model for talking head translation, TransFace, which can translate audio-visual speech into audio-visual speech in other languages. |
| Outcome: | The proposed model can translate audio-visual speech into audio-visual speech in other languages. |
The Butterfly Effect of Model Editing: Few Edits Can Trigger Large Language Models Collapse (2024.findings-acl)
Copied to clipboard
| Challenge: | Even a single edit can trigger model collapse, manifesting as significant performance degradation in various benchmark tasks. |
| Approach: | They propose to use perplexity as a surrogate metric to determine whether an edited model's performance is affected by a single edit. |
| Outcome: | The proposed method shows that even a single edit can cause model collapse, manifesting as significant performance degradation in various benchmark tasks. |
Knowledge Graph Retrieval-Augmented Generation for LLM-based Recommendation (2025.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have produced significant advances in the field of recommender systems. |
| Approach: | They propose to retrieve up-to-date structure information from the knowledge graph to augment recommendations by leveraging external knowledge sources. |
| Outcome: | Experiments on a large dataset show that the proposed method is effective in enhancing LLM-based recommendations. |
AutoVecCoder: Teaching LLMs to Generate Explicitly Vectorized Code (2026.findings-acl)
Copied to clipboard
ShangZhan Li, Xinyu Yin, Xuanyu Jin, Ye He, Yuxin Zhou, Yuxuan Li, Xu Han, Wanxiang Che, Qi Shi, Ting Liu, Maosong Sun
| Challenge: | Current development practices face a dichotomy between automation and performance. |
| Approach: | They propose a framework to empower LLMs with the capability of automated explicit vectorization. |
| Outcome: | The proposed framework achieves state-of-the-art performance on the SSE and AVX subsets of SimdBench. |
DAMON: A Dialogue-Aware MCTS Framework for Jailbreaking Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for multi-turn attacks mainly utilize a predefined dialogue pattern, limiting their effectiveness in realistic situations. |
| Approach: | They propose a multi-turn jailbreak attack method that leverages Monte Carlo Tree Search to explore multi-turned conversational spaces and identifies sub-instruction sequences that induce harmful responses. |
| Outcome: | The proposed method can induce undesired behaviors across five LLMs and three datasets. |
LEDOM: Reverse Language Model (2026.acl-long)
Copied to clipboard
Xunjian Yin, Sitao Cheng, Yuxi Xie, Xinyu Hu, Li Lin, Xinyi Wang, Liangming Pan, William Yang Wang, Xiaojun Wan
| Challenge: | Autoregressive language models are trained exclusively left-to-right, yet they are limited in their ability to factorize text. |
| Approach: | They propose a purely reverse autoregressive language model that factorizes text as a product of left-to-right conditionals. |
| Outcome: | The proposed model can be used to score forward outputs using reverse posterior estimates. |
MATCH: Modulating Attention via In-Context Retrieval for Long-Context Transformers (2026.acl-long)
Copied to clipboard
Linrui Ma, Chun Hei Lo, Xinyu Wang, Peng Lu, Xihao Yuan, Hanting Chen, Kai Han, Xinghao Chen, Chengjun Zhan, Hanlin xu, Yichun Yin, Lifeng Shang, Feng Wen, Boxing Chen, Yufei Cui
| Challenge: | Existing approaches to improve efficiency often enforce rigid structural constraints such as local attention windows. |
| Approach: | They propose a framework that augments sparse-attention mechanisms with dynamically integrated in-context information through an efficient retrieval system. |
| Outcome: | Empirical results show that MATCH significantly improves the performance of sparse-attention models on synthetic and real-world natural-language tasks. |
The Fall of ROME: Understanding the Collapse of LLMs in Model Editing (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies have found that model editing methods can cause large language models to collapse with just a single edit. |
| Approach: | They propose a method that uses prefixed keys and adds prefixes during testing to prevent model collapse. |
| Outcome: | The proposed method prevents model collapse while maintaining effectiveness, the authors show . Rank-One Model Editing (ROME) has been found to cause model collapse with just a single edit . |
MAIR: A Massive Benchmark for Evaluating Instructed Retrieval (2024.emnlp-main)
Copied to clipboard
Weiwei Sun, Zhengliang Shi, Wu Long, Lingyong Yan, Xinyu Ma, Yiding Liu, Min Cao, Dawei Yin, Zhaochun Ren
| Challenge: | Existing IR benchmarks focus on a limited scope of tasks, making them insufficient for evaluating the latest IR models. |
| Approach: | They propose a multi-task instruction-tuned IR benchmark that includes 126 distinct IR tasks across 6 domains. |
| Outcome: | The proposed model performs better on instruction-tuned models than non-instruction-tunned models on MAIR. |