When Teachers Mislead: Spurious-Signal-Aware On-Policy Distillation Paper • 2608.03632 • Published 12 days ago • 23
SwanTale: Unified Multi-Speaker Speech and Audio Generation for Instruct and Zero-Shot Tasks Paper • 2608.02023 • Published 13 days ago • 155
DecoupleMix: Decoupled Ratio Search and Convex Allocation for Scalable VLM Data Recipes Paper • 2607.24516 • Published 20 days ago • 8
TableVerse: A Large-scale Tabletop Dataset with Real-world Grounded Layouts for Generalizable Manipulation Paper • 2607.21017 • Published 24 days ago • 8
SWE-Pruner Pro: The Coder LLM Already Knows What to Prune Paper • 2607.18213 • Published 27 days ago • 79
UniVR: Thinking in Visual Space for Unified Visual Reasoning Paper • 2607.12800 • Published Jul 14 • 32
ShortOPD: Recovering Pruned LLMs with Short-to-Long On-Policy Distillation Paper • 2607.13124 • Published Jul 14 • 20
SafePyramid: A Hierarchical Benchmark for In-context Policy Guardrailing Paper • 2606.29887 • Published Jun 29 • 6