Logit-Contribution Scoring Identifies Non-Literal Retrieval Heads Paper • 2607.01002 • Published 13 days ago • 19
Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent Paper • 2606.30616 • Published 15 days ago • 98
VISTA: View-Consistent Self-Verified Training for GUI Grounding Paper • 2606.14579 • Published Jun 12 • 8
LLaVA-OneVision-2: Towards Next-Generation Perceptual Intelligence Paper • 2605.25979 • Published May 25 • 27
DelTA: Discriminative Token Credit Assignment for Reinforcement Learning from Verifiable Rewards Paper • 2605.21467 • Published May 20 • 207
IndusAgent: Reinforcing Open-Vocabulary Industrial Anomaly Detection with Agentic Tools Paper • 2605.20682 • Published May 20 • 85
Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information Paper • 2605.11609 • Published May 12 • 196