| 2026-10-22 | 2610.17860 | Socratic Skill Learning: Interleaving Teaching and Practice for Transferable Agent Skills | Letian Zhu et al. | arXiv | 45 | |
| 2026-10-09 | 2610.12289 | TestPrism: Rethinking Test Evaluation Beyond a Single Reference | Han Li et al. | arXiv | 35 | |
| 2026-08-06 | 2607.24821 | AVE-Compass: Towards Holistic Evaluation for Audio-Video Editing Abilities | Yuqing Wen et al. | arXiv | 11 | |
| 2026-06-23 | 2606.22883 | CLI-Universe: Towards Verifiable Task Synthesis Engine for Terminal Agents | Zhanbo Hua et al. | arXiv | 38 | |
| 2026-06-09 | 2606.08415 | CoVEBench: Can Video Editing Models Handle Complex Instructions? | Jiangtao Wu et al. | arXiv | 47 | |
| 2026-06-09 | 2606.08572 | OmniCap-IF: Benchmarking and Improving Instruction Following Abilities for Omni-Video Captioning | Jiahao Wang et al. | arXiv | 13 | |
| 2026-06-04 | 2606.02060 | Where Do Deep-Research Agents Go Wrong? Span-Level Error Localization in Agent Trajectories | Jiaming Wang et al. | arXiv | 58 | |
| 2026-06-04 | 2606.01993 | MMG2Skill: Can Agents Distill In-the-Wild Guides into Self-Evolving Skills? | Xinyu Che et al. | arXiv | 16 | |
| 2026-06-02 | 2606.02320 | TVIR: Building Deep Research Agents Towards Text--Visual Interleaved Report Generation | Xinkai Ma et al. | arXiv | 11 | |
| 2026-05-18 | 2605.15301 | Solvita: Enhancing Large Language Models for Competitive Programming via Agentic Evolution | Han Li et al. | arXiv | 21 | |