| 2026-10-20 | 2610.17724 | Evidence-First Agentic Verification | Reza Sanayei et al. | arXiv | 21 | |
| 2026-10-12 | 2610.08966 | Humanity's Sixth Sense: Benchmarking Intuitive Visual Reasoning in Multimodal Models | Xingang Guo et al. | arXiv | 12 | |
| 2026-09-21 | 2609.12623 | SteerDuplex: Steerable Duplex Speech Dialogue Models | Utkarsh Tyagi et al. | arXiv | 5 | |
| 2026-09-15 | 2609.13463 | Root-Cause Attribution Is a Search Problem: Continual Search for Long-Horizon Agent Failures | Harsh Raj et al. | arXiv | 5 | |
| 2026-09-14 | 2609.10824 | Studying Without a Syllabus: Task-Agnostic Environment Preprocessing | Vinay Samuel et al. | arXiv | 5 | |
| 2026-08-07 | 2608.06301 | HarnessOpt-Bench: Evaluating LLMs at Harness Optimization | Varun Ursekar et al. | arXiv | 15 | |
| 2026-08-04 | 2607.28802 | Model or Harness? An Interaction-Centric Taxonomy for Localizing Agent Failures | Harsh Raj et al. | arXiv | 9 | |
| 2026-07-01 | 2606.30573 | SWE-INTERACT: Reimagining SWE Benchmarks as User-Driven Long-Horizon Coding Sessions | Mohit Raghavendra et al. | arXiv | 7 | |
| 2026-05-20 | 2605.20164 | Not Every Rubric Teaches Equally: Policy-Aware Rubric Rewards for RLVR | Utkarsh Tyagi et al. | arXiv | 5 | |
| 2026-05-05 | 2604.09408 | HiL-Bench (Human-in-Loop Benchmark): Do Agents Know When to Ask for Help? | Mohamed Elfeki et al. | arXiv | 4 | |