SEKE研究组SEKE研究组
EvoCoT论文解读——ACL 2026论文 Blog
RLVR已成为提升大语言模型推理能力的重要方法，但在困难问题上，由于模型初始采样准确率较低，奖励信号稀疏，导致模型难以有效探索。现有方法通常依赖教师模型生成推理轨迹或过滤困难样本，…
最新论文全部 →
arXiv 论文
- [arXiv 2026] Yihong Dong, Jianha Xiao, Xue Jiang, Xuyuan Guo, Zhiyuan Fan, Jiaru Qian, Kechi Zhang, Jia Li, Zhi Jin, Ge Li; Evaluating the Formal Reasoning Capabilities of Large Language Models through Chomsky Hierarchy; arXiv preprint, arXiv:2604.02709, 2026. PDF
- [arXiv 2026] Huanyu Liu, Ge Li, Yihong Dong, Sihan Wu, Peixu Wang, Sihao Cheng, Taozhi Chen, Kechi Zhang, Hao Zhu, Tongxuan Liu; Do Transformers Have the Ability for Periodicity Generalization?; arXiv preprint, arXiv:2601.22690, 2026. PDF
已发表论文
- [软件学报, 2026] Yihong Dong, Xue Jiang, Jiaru Qian, Tian Wang, Kechi Zhang, Zhi Jin, Ge Li; 基于大语言模型智能体的代码生成综述（A Survey on Code Generation with LLM-based Agents）; 软件学报 (软件学报); Vol. 37, Iss. 8, pp.3089-3115. PDF
- [AAAI 2026] Xue Jiang, Yihong Dong, Zheng Fang, Yingwei Ma, Tangxinyu Wang, Rongyu Cao, Binhua Li, Zhi Jin, Wenpin Jiao, Yongbin Li, Ge Li; Large Language Model Unlearning for Source Code; The 40th Annual AAAI Conference on Artificial Intelligence (AAAI 2026), Singapore, Singapore. PDF
- [ISSTA 2026] Yongmin Li, Yihong Dong, Jia Li, Ge Li; Efficient Grammar-Constrained Decoding via Parser Stack Classification; The ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA 2026), Oakland, California, USA, October 4 - 9, 2026.
- [SCIS, 2026] Fang Liu, Ge Li, Qianhui Zhao, Li Zhang; Learning to Represent Code Semantics; Science China Information Sciences (SCIS); Vol. 68, Iss. 7, pp.172101:1-172101:21.
- [ACL 2026] Yihong Dong, Xue Jiang, Yongding Tao, Huanyu Liu, Kechi Zhang, Lili Mou, Rongyu Cao, Yingwei Ma, Jue Chen, Binhua Li, Zhi Jin, Fei Huang, Yongbin Li, Ge Li; Rl-plus: Countering Capability Boundary Collapse of LLMs in Reinforcement Learning with Hybrid-policy Optimization; Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), San Diego, California, USA, July 2 - 7, 2026. PDF
- [ACL 2026] Yihong Dong, Zhaoyu Ma, Xue Jiang, Zhiyuan Fan, Jiaru Qian, Yongmin Li, Jianha Xiao, Zhi Jin, Rongyu Cao, Binhua Li, Fei Huang, Yongbin Li, Ge Li; Saber: Efficient Sampling with Adaptive Acceleration and Backtracking Enhanced Remasking for Diffusion Language Model in Code Generation; Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), San Diego, California, USA.
- [FSE 2026] Yuqi Zhu, Ge Li, Hong Mei, Zhi Jin, Jia Li, Qibin Zheng, Jieyuan Zhang; Influence-Aware Bayesian-Inspired Token Reweighting for Improved Code Generation; Proceedings of the ACM International Conference on the Foundations of Software Engineering (FSE 2026), Montreal, Canada.
- [FSE 2026] Hao Zhu, Jia Li, Cuiyun Gao, Jiaru Qian, Yihong Dong, Huanyu Liu, Lecheng Wang, Ziliang Wang, Xiaolong Hu, Ge Li; VulInstruct: Teaching LLMs Root-Cause Reasoning for Vulnerability Detection via Security Specifications; Proceedings of the ACM International Conference on the Foundations of Software Engineering (FSE 2026), Montreal, Canada.
- [ICSE 2026] Yongmin Li, Jia Li, Ge Li, Zhi Jin; AdapTrack: Constrained Decoding without Distorting LLM's Output Intent; Proceedings of the 48th IEEE/ACM International Conference on Software Engineering (ICSE 2026), Rio de Janeiro, Brazil.
- [ICSE 2026] Kechi Zhang, Huangzhao Zhang, Ge Li, Jinliang You, Jia Li, Yunfei Zhao, Zhi Jin; SEAlign: Alignment Training for Software Engineering Agent; Proceedings of the 48th IEEE/ACM International Conference on Software Engineering (ICSE 2026), Rio de Janeiro, Brazil. PDF
- [ACL 2026] Huanyu Liu, Jia Li, Chang Yu, Taozhi Chen, Yihong Dong, Lecheng Wang, Yongding Tao, Xiaolong Hu, Ge Li; EvoCoT: Overcoming the Exploration Bottleneck in Reinforcement Learning; Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), San Diego, California, USA. PDF
Blog → - [ICLR 2026] Yongding Tao, Tian Wang, Yihong Dong, Huanyu Liu, Kechi Zhang, Xiaolong Hu, Ge Li; Detecting Data Contamination from Reinforcement Learning Post-training for Large Language Models; Proceedings of the Fourteenth International Conference on Learning Representations (ICLR 2026), Rio de Janeiro, Brazil. PDF
- [ACL 2026] Xue Jiang, Yihong Dong, Mengyang Liu, Hongyi Deng, Tian Wang, Yongding Tao, Rongyu Cao, Binhua Li, Zhi Jin, Wenpin Jiao, Fei Huang, Yongbin Li, Ge Li; CODERL+: Improving Code Generation via Reinforcement with Execution Semantics Alignment; Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), San Diego, California, USA. PDF
- [ACL 2026] Zheng Fang, Yihong Dong, Lili Mou, Dongming Jin, Zhi Jin, Ge Li; IntentCoding: Amplifying User Intent in Code Generation; Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), San Diego, California, USA. PDF
- [ACL 2026] Xue Jiang, Ge Li, Jiaru Qian, Xianjie Shi, Chenjie Li, Hao Zhu, Ziyu Wang, Jielun Zhang, Zheyu Zhao, Lingwei Wu, Kechi Zhang, Jia Li, Wenpin Jiao, Zhi Jin, Yihong Dong; KOCO-BENCH: Can Large Language Models Leverage Domain Knowledge in Software Development?; Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), San Diego, California, USA. PDF
- [TOSEM, 2025] Jia Li, Ge Li, Yongmin Li, Jin Zhi; Structured Chain-of-Thought Prompting for Code Generation; ACM Transactions on Software Engineering and Methodology (TOSEM); Vol. 34, Iss. 2, pp.1-23.
- [TOSEM, 2025] Yuwei Zhang, Zhi Jin, Ying Xing, Ge Li, Fang Liu, Jiaxin Zhu, Wensheng Dou, Jun Wei; PATCH: Empowering Large Language Model with Programmer-Intent Guidance and Collaborative-Behavior Simulation for Automatic Bug Fixing; ACM Transactions on Software Engineering and Methodology.
- [EMSE, 2025] Jia Li, Zheng Fang, Xianjie Shi, Zhi Jin, Fang Liu, Yunfei Zhao, Ge Li; SCodeSearcher: Soft Contrastive Learning for Code Search; Empirical Software Engineering (EMSE); Vol. 30, Iss. 3, pp.1-23.