| 2026-10-16 | 2610.16784 | InfoVis: Unified Slide Image Generation and Editing | Jiahao Wang et al. | arXiv | 32 | |
| 2026-10-01 | 2609.38288 | AREX-2: Advancing Self-Improving Agents through Long-Horizon Reflective Tasks | Hongjin Qian et al. | arXiv | 92 | |
| 2026-09-29 | 2609.32712 | MassAlloc Attention: Let Attention Allocate Its Own Compute | Jingze Shi et al. | arXiv | 42 | |
| 2026-09-29 | 2609.32704 | CoWindow Attention: Full Causal Coverage Is a Collective Property | Jingze Shi et al. | arXiv | 32 | |
| 2026-09-03 | 2609.02749 | Repo-To-Skill: Distilling GitHub Repositories Into AI4AI Skills | Jianlyu Chen et al. | arXiv | 137 | |
| 2026-08-21 | 2608.20281 | Inject, Align, Recover: Staged Post-Training for Retrieval-Free Document Knowledge Internalization | Qian Kou et al. | arXiv | 10 | |
| 2026-08-04 | 2608.01862 | Wnuan: Staged Post-Training for Question Answering over Proprietary Enterprise Knowledge | Xiaofeng Shi et al. | arXiv | 6 | |
| 2026-07-24 | 2607.21461 | AREX: Towards a Recursively Self-Improving Agent for Deep Research | Shuqi Lu et al. | arXiv | 95 | |
| 2026-06-25 | 2606.20023 | When Lower Privileges Suffice: Investigating Over-Privileged Tool Selection in LLM Agents | Kaiyue Yang et al. | arXiv | 3 | |
| 2026-06-24 | 2606.23997 | ChartWalker: Benchmarking the Cross-Chart RAG Task | Ning Tang et al. | arXiv | 4 | |
| 2026-05-01 | 2604.27711 | ExoActor: Exocentric Video Generation as Generalizable Interactive Humanoid Control | Yanghao Zhou et al. | arXiv | 39 | |