Submitted by akshaynambi 9 Magentic Marketplace: An Open-Source Environment for Studying Agentic Markets Microsoft Research 21 2
Submitted by lynazhang 59 LoongRL:Reinforcement Learning for Advanced Reasoning over Long Contexts Microsoft Research 5
Submitted by jeepliu 26 DocReward: A Document Reward Model for Structuring and Stylizing Microsoft Research 3
Submitted by martinagvilas 1 Tracing the Traces: Latent Temporal Signals for Efficient and Accurate Reasoning Microsoft Research 2
Submitted by zli999 4 PixelCraft: A Multi-Agent System for High-Fidelity Visual Reasoning on Structured Images Microsoft Research 11 2
Submitted by yshenaw 9 Benefits and Pitfalls of Reinforcement Learning for Language Model Planning: A Theoretical Perspective Microsoft Research 2