Daily papers of 27 Oct 2026

Daily Papers

Here is the selection of papers for today (27 Oct):

• Trajectory2Gym: Decision-Grounded Compositional Interaction Synthesis for Tool-Use Agents (336 ▲)
• Trial by Trace: Benchmarking AI Agent Harnesses for Reliability and Failure Attribution (249 ▲)
• StreamVibe: Proactive Video Agents for Evolving User Intent (245 ▲)
• DynaFlow: Learning Free-Form Multi-Agent Orchestration (189 ▲)
• InteractiveWorld: A Verifiable Human-Centric Environment for Building Better Interactive MLLM Assistants (113 ▲)
• Work Agents: Systems, Capabilities, and Optimization (103 ▲)
• Continual Learning Exam: Benchmarking Long-Horizon RSI and Experiential Learning in LLM Agents (49 ▲)
• WorldCast: World-Consistent Character Video Generation with Large-Angle Camera Control (46 ▲)
• Spatial Audio Generation and Editing with Executable Plans and Self-Distilling Transfusion (42 ▲)
• AV-FABLE: Audio-Visual Manipulation Forensics in Long-Form Videos (37 ▲)
• Eat Your Precision: Precision Objective Mitigates Reward Hacking in Rubric-based RL (33 ▲)
• Principia: Test-Time Hypotheses to Principle Construction for Scientific Discovery (27 ▲)
• And 49 more papers

NEW - I'm opening Daily Papers! If you're a researcher you can now submit your peers' papers to be featured in Daily Papers directly. Use this page to do so: huggingface.co/papers/submit.

Keep exploring,
AK and the research community