Confidence Comes from Experience: Experiential Confidence Estimation from Reasoning to Agents Paper • 2609.17708 • Published 6 days ago • 56
OSReward: Instituting Standardized Evaluation for Cross-Platform Computer-Use Reward Models Paper • 2607.28609 • Published Jul 30 • 74
From Proprietary to Open-Source: Bridging the Distribution Gap via Multi-Agent Protocol Distillation in Agentic Search Paper • 2607.24280 • Published Jul 27 • 85
WBench: A Comprehensive Multi-turn Benchmark for Interactive Video World Model Evaluation Paper • 2605.25874 • Published May 25 • 82
LLMEval-Logic: A Solver-Verified Chinese Benchmark for Logical Reasoning of LLMs with Adversarial Hardening Paper • 2605.19597 • Published May 19 • 20