WBench: A Comprehensive Multi-turn Benchmark for Interactive Video World Model Evaluation Paper • 2605.25874 • Published 5 days ago • 97
Forecasting Scientific Progress with Artificial Intelligence Paper • 2605.22681 • Published 9 days ago • 42
MixSD: Mixed Contextual Self-Distillation for Knowledge Injection Paper • 2605.16865 • Published 14 days ago • 7
ESI-Bench: Towards Embodied Spatial Intelligence that Closes the Perception-Action Loop Paper • 2605.18746 • Published 12 days ago • 5
From Context to Skills: Can Language Models Learn from Context Skillfully? Paper • 2604.27660 • Published 27 days ago • 166
backuppp/lirpg-fullparam-qwen2-5-math-7b-answeronly01-handrolled-zeroinit-gn-lrin5e-5-nostd-nokl Updated 28 days ago • 1
The Master Key Hypothesis: Unlocking Cross-Model Capability Transfer via Linear Subspace Alignment Paper • 2604.06377 • Published Apr 7 • 7
Rethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model Capability Paper • 2604.06628 • Published Apr 8 • 326
Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding Paper • 2604.05015 • Published Apr 6 • 236