Microsoft AI

Today’s Open Source (2025-10-10): Microsoft Releases UserLM for User Role Simulation in Conversations, Advancing Real Interaction Technology

open source AI

Today’s Open Source (2025-10-10): Microsoft Releases UserLM for User Role Simulation in Conversations, Advancing Real Interaction Technology

Daily Discovery of Latest LLMs Date: 2025-10-10 · Location: Hong Kong, China --- 📢 Overview Highlighted Releases: * Language Model: UserLM * Foundation Model: Lumina-DiMOO * Language Model (Code): CoDA-v0-Instruct * Reasoning Model: Jamba-Reasoning * Visualization Tool: Model Explorer ONNX * Video Framework: Code2Video --- --- 🏆 Foundation Models 1. UserLM — Simulating the User Side of Conversations Key Points:

By Honghao Wang
Any Agent Can Use Reinforcement Learning: Microsoft Launches Agent Lightning Framework with No Code Changes

Agent Lightning

Any Agent Can Use Reinforcement Learning: Microsoft Launches Agent Lightning Framework with No Code Changes

# **微软推出 Agent Lightning:为任意 AI Agent 提供可扩展的强化学习训练框架** > **新智元导读** > AI Agent 已逐渐从科幻步入现实:不仅可编写代码、调用工具、进行多轮对话,还能端到端完成软件开发,广泛应用于金融、游戏与软件工程。 > 然而,当前训练与优化环节仍面临挑战,传统强化学习在复杂动态交互场景中表现欠佳。 > 微软提出 **Agent Lightning** —— 一个灵活、可扩展的框架,可对任意 AI Agent 进行基于强化学习的 LLM 训练,并已在 arXiv 发表研究论文。 > [论文链接](https://arxiv.org/abs/2508.03680) --- ## 📌 **核心贡献** 1. **训练与

By Honghao Wang