2026

CVE-Factory: Scaling Expert-Level Agentic Tasks for Code Security Vulnerability

Proceedings of the 43rd International Conference on Machine Learning, 2026.

Luo, Xianzhen and Zhang, Jingyuan and Zhou, Shiqi and Huang, Jinyang and Xiao, Chuan and Zhu, Qingfu and Ma, Zhiyuan and Yue, Xing and Yue, Yang and Zeng, Wencong and others

CVE-Factory: Scaling Expert-Level Agentic Tasks for Code Security Vulnerability

Proceedings of the 43rd International Conference on Machine Learning, 2026.

Luo, Xianzhen and Zhang, Jingyuan and Zhou, Shiqi and Huang, Jinyang and Xiao, Chuan and Zhu, Qingfu and Ma, Zhiyuan and Yue, Xing and Yue, Yang and Zeng, Wencong and others

Know more, know clearer: A meta-cognitive framework for knowledge augmentation in large language models

Proceedings of the 43rd International Conference on Machine Learning, 2026.

Chen, Hao and He, Ye and Fan, Yuchun and Yan, Yukun and Liu, Zhenghao and Zhu, Qingfu and Sun, Maosong and Che, Wanxiang

Know more, know clearer: A meta-cognitive framework for knowledge augmentation in large language models

Proceedings of the 43rd International Conference on Machine Learning, 2026.

Chen, Hao and He, Ye and Fan, Yuchun and Yan, Yukun and Liu, Zhenghao and Zhu, Qingfu and Sun, Maosong and Che, Wanxiang

MEnvAgent: Scalable Polyglot Environment Construction for Verifiable Software Engineering

Proceedings of the 43rd International Conference on Machine Learning, 2026.

Guo, Chuanzhe and Wu, Jingjing and He, Sijun and Chen, Yang and Kuang, Zhaoqi and Fan, Shilong and Chen, Bingjin and Bao, Siqi and Liu, Jing and Wu, Hua and others

MEnvAgent: Scalable Polyglot Environment Construction for Verifiable Software Engineering

Proceedings of the 43rd International Conference on Machine Learning, 2026.

Guo, Chuanzhe and Wu, Jingjing and He, Sijun and Chen, Yang and Kuang, Zhaoqi and Fan, Shilong and Chen, Bingjin and Bao, Siqi and Liu, Jing and Wu, Hua and others

AutoVecCoder: Teaching LLMs to Generate Explicitly Vectorized Code

Findings of the Association for Computational Linguistics: ACL 2026, 31942--31959, 2026.

Li, ShangZhan and Yin, Xinyu and Jin, Xuanyu and He, Ye and Zhou, Yuxin and Li, Yuxuan and Han, Xu and Che, Wanxiang and Shi, Qi and Liu, Ting and Sun, Maosong

AutoVecCoder: Teaching LLMs to Generate Explicitly Vectorized Code

Findings of the Association for Computational Linguistics: ACL 2026, 31942--31959, 2026.

Li, ShangZhan and Yin, Xinyu and Jin, Xuanyu and He, Ye and Zhou, Yuxin and Li, Yuxuan and Han, Xu and Che, Wanxiang and Shi, Qi and Liu, Ting and Sun, Maosong

Dashboard2Code: Evaluating Multimodal Models on Reconstructing Interactive Dashboards

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 37696--37732, 2026.

Niu, Tianhao and Han, Ziyu and Chen, Qiguang and Zhou, Shiqi and Shan, Baocai and Fang, Hengjie and Zhu, Qingfu and Che, Wanxiang

Dashboard2Code: Evaluating Multimodal Models on Reconstructing Interactive Dashboards

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 37696--37732, 2026.

Niu, Tianhao and Han, Ziyu and Chen, Qiguang and Zhou, Shiqi and Shan, Baocai and Fang, Hengjie and Zhu, Qingfu and Che, Wanxiang

Format-Adapter: Improving Reasoning Capability of LLMs by Adapting Suitable Format

Findings of the Association for Computational Linguistics: ACL 2026, 22408--22427, 2026.

Wang, Dingzirui and Zhang, Xuanliang and Cao, Rongyu and Dou, Longxu and Luo, Xianzhen and Ma, Yingwei and Zhu, Qingfu and Che, Wanxiang and Li, Binhua and Huang, Fei and others

Format-Adapter: Improving Reasoning Capability of LLMs by Adapting Suitable Format

Findings of the Association for Computational Linguistics: ACL 2026, 22408--22427, 2026.

Wang, Dingzirui and Zhang, Xuanliang and Cao, Rongyu and Dou, Longxu and Luo, Xianzhen and Ma, Yingwei and Zhu, Qingfu and Che, Wanxiang and Li, Binhua and Huang, Fei and others

OMIBench: Benchmarking Olympiad-Level Multi-Image Reasoning in Large Vision-Language Models

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 45100--45135, 2026.

Chen, Qiguang and Luan, Chengyu and Wu, Jiajun and Yu, Qiming and Yang, Yi and Li, Yizhuo and Tong, Jingqi and Feng, Xiachong and Qin, Libo and Che, Wanxiang

OMIBench: Benchmarking Olympiad-Level Multi-Image Reasoning in Large Vision-Language Models

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 45100--45135, 2026.

Chen, Qiguang and Luan, Chengyu and Wu, Jiajun and Yu, Qiming and Yang, Yi and Li, Yizhuo and Tong, Jingqi and Feng, Xiachong and Qin, Libo and Che, Wanxiang

Scaling laws for code: A more data-hungry regime

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 24005--24021, 2026.

Luo, Xianzhen and Zheng, Wenzhen and Zhu, Qingfu and Zhang, Rongyi and Li, Houyi and Huang, Siming and Fan, YuanTao and Che, Wanxiang

Scaling laws for code: A more data-hungry regime

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 24005--24021, 2026.

Luo, Xianzhen and Zheng, Wenzhen and Zhu, Qingfu and Zhang, Rongyi and Li, Houyi and Huang, Siming and Fan, YuanTao and Che, Wanxiang

Seer Self-Consistency: Advance Budget Estimation for Adaptive Test-Time Scaling

Findings of the Association for Computational Linguistics: ACL 2026, 42734--42747, 2026.

Ji, Shiyu and Wang, Yixuan and Liu, Yijun and Zhu, Qingfu and Che, Wanxiang

Seer Self-Consistency: Advance Budget Estimation for Adaptive Test-Time Scaling

Findings of the Association for Computational Linguistics: ACL 2026, 42734--42747, 2026.

Ji, Shiyu and Wang, Yixuan and Liu, Yijun and Zhu, Qingfu and Che, Wanxiang

When Does Language Matter? Multilingual Instructions Reveal Step-wise Language Sensitivity in Vision-Language-Action Models

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 44615--44629, 2026.

Dong, Xuan and Han, Zhe and Niu, Tianhao and Zhu, Qingfu and Che, Wanxiang

When Does Language Matter? Multilingual Instructions Reveal Step-wise Language Sensitivity in Vision-Language-Action Models

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 44615--44629, 2026.

Dong, Xuan and Han, Zhe and Niu, Tianhao and Zhu, Qingfu and Che, Wanxiang

Bounds of Chain-of-Thought Robustness: Reasoning Steps, Embed Norms, and Beyond

The Fourteenth International Conference on Learning Representations, 2026.

Wang, Dingzirui and Zhang, Xuanliang and Xu, Keyan and Zhu, Qingfu and Che, Wanxiang and Deng, Yang

Bounds of Chain-of-Thought Robustness: Reasoning Steps, Embed Norms, and Beyond

The Fourteenth International Conference on Learning Representations, 2026.

Wang, Dingzirui and Zhang, Xuanliang and Xu, Keyan and Zhu, Qingfu and Che, Wanxiang and Deng, Yang

How Many Code and Test Cases Are Enough? Evaluating Test Cases Generation from a Binary-Matrix Perspective

The Fourteenth International Conference on Learning Representations, 2026.

Luo, Xianzhen and Huang, Jinyang and Zheng, Wenzhen and Zhu, Qingfu and Xu, Mingzheng and Xu, Yiheng and Fan, Yuantao and Qin, Libo and Che, Wanxiang

How Many Code and Test Cases Are Enough? Evaluating Test Cases Generation from a Binary-Matrix Perspective

The Fourteenth International Conference on Learning Representations, 2026.

Luo, Xianzhen and Huang, Jinyang and Zheng, Wenzhen and Zhu, Qingfu and Xu, Mingzheng and Xu, Yiheng and Fan, Yuantao and Qin, Libo and Che, Wanxiang

ProxyAttn: Guided Sparse Attention via Representative Heads

The Fourteenth International Conference on Learning Representations, 2026.

Wang, Yixuan and He, Huang and Bao, Siqi and Wu, Hua and Wang, Haifeng and Zhu, Qingfu and Che, Wanxiang

ProxyAttn: Guided Sparse Attention via Representative Heads

The Fourteenth International Conference on Learning Representations, 2026.

Wang, Yixuan and He, Huang and Bao, Siqi and Wu, Hua and Wang, Haifeng and Zhu, Qingfu and Che, Wanxiang

Aware First, Think Less: Dynamic Boundary Self-Awareness Drives Significant Gains in Reasoning Efficiency in Large Language Models

Proceedings of the AAAI Conference on Artificial Intelligence, 40(36), 30261--30269, 2026.

Chen, Qiguang and Peng, Dengyun and Liu, Jinhao and Su, HuiKang and Guan, Jiannan and Qin, Libo and Che, Wanxiang

Aware First, Think Less: Dynamic Boundary Self-Awareness Drives Significant Gains in Reasoning Efficiency in Large Language Models

Proceedings of the AAAI Conference on Artificial Intelligence, 40(36), 30261--30269, 2026.

Chen, Qiguang and Peng, Dengyun and Liu, Jinhao and Su, HuiKang and Guan, Jiannan and Qin, Libo and Che, Wanxiang

CAMERA: Multi-Matrix Joint Compression for MoE Models via Micro-Expert Redundancy Analysis

Proceedings of the AAAI Conference on Artificial Intelligence, 40(32), 27395--27404, 2026.

Xu, Yuzhuang and Han, Xu and Zhang, Yuanchi and Wang, Yixuan and Liu, Yijun and Ji, Shiyu and Zhu, Qingfu and Che, Wanxiang

CAMERA: Multi-Matrix Joint Compression for MoE Models via Micro-Expert Redundancy Analysis

Proceedings of the AAAI Conference on Artificial Intelligence, 40(32), 27395--27404, 2026.

Xu, Yuzhuang and Han, Xu and Zhang, Yuanchi and Wang, Yixuan and Liu, Yijun and Ji, Shiyu and Zhu, Qingfu and Che, Wanxiang

Judge Q: Trainable Queries for Optimized Information Retention in KV Cache Eviction

Proceedings of the AAAI Conference on Artificial Intelligence, 40(38), 32240--32248, 2026.

Liu, Yijun and Wang, Yixuan and Xu, Yuzhuang and Ji, Shiyu and Xu, Yang and Zhu, Qingfu and Che, Wanxiang

Judge Q: Trainable Queries for Optimized Information Retention in KV Cache Eviction

Proceedings of the AAAI Conference on Artificial Intelligence, 40(38), 32240--32248, 2026.

Liu, Yijun and Wang, Yixuan and Xu, Yuzhuang and Ji, Shiyu and Xu, Yang and Zhu, Qingfu and Che, Wanxiang