Beyond Domain-Routed Supervision: Cross-Teacher Guarding for Multi-Teacher On-Policy Distillation
Paper • 2610.13521 • Published • 243
Daniel Martin
vibemartin
AI & ML interests
None yet
Recent Activity
View all activity
Organizations
None yet
upvoted 7
papers 1 day ago
Any Time, Any Space: A World-State-Centric Framework for Agentic Video Generation
Paper • 2610.20571 • Published • 343
Player-1: Learning to Play from Synthetic Experience
Paper • 2610.20584 • Published • 334
Speculative Forcing for Efficient Autoregressive Video Generation
Paper • 2610.22347 • Published • 76
LoopVL: Recurrent Visual Intelligence
Paper • 2609.38426 • Published • 485
The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation
Paper • 2609.36484 • Published • 564
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents
Paper • 2609.39102 • Published • 665
upvoted 8
papers 15 days ago
PROWBench: Do Video Models Render What the Program Specifies?
Paper • 2610.02205 • Published • 75
PhysVista: Benchmarking Physical Intelligence in VLMs via a Perception-Reasoning-Assessment Loop
Paper • 2610.00559 • Published • 69
World Observer: Joint Actor-Observer Generation for Persistent World Modeling
Paper • 2610.02162 • Published • 91
MolmoAct2: Action Reasoning Models for Real-world Deployment
Paper • 2605.02881 • Published • 359
Orca: The World is in Your Mind
Paper • 2606.30534 • Published • 514
StateM: Reaching 95.3% Raw Accuracy, or a \$15 Frontier Run, on Terminal-Bench 2.1 via Harness Scaling
Paper • 2608.15089 • Published • 452
StudentSim: Training LLM-based Student Simulators
Paper • 2609.01591 • Published • 611
LimiX-2: A Contextual Mechanism Network Towards General Structured-Data Intelligence
Paper • 2609.17488 • Published • 865