Snowflake 的智能体强化学习训练方法

🚀 Snowflake AI Research is doubling down on 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗥𝗲𝗶𝗻𝗳𝗼𝗿𝗰𝗲𝗺𝗲𝗻𝘁 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 (𝗥𝗟). Training long-horizon agents requires more than throwing compute at the problem—it demands finding the right recipe, eliminating reward hacking, and co-designing systems with modeling. Today, we’re sharing three milestones highlighting our end-to-end approach: 1️⃣ 𝗙𝗶𝗻𝗱 𝘁𝗵𝗲 𝗥𝗲𝗰𝗶𝗽𝗲 (𝗔𝗿𝗰𝘁𝗶𝗰-𝗦𝗪𝗘): 𝘖𝘷𝘦𝘳𝘧𝘪𝘵 𝘧𝘪𝘳𝘴𝘵, 𝘵𝘩𝘦𝘯 𝘴𝘤𝘢𝘭𝘦. Before scaling to thousands of tasks, we ensure our RL pipeline can cleanly overfit a 160-task sandbox and hold it without collapsing. If a recipe leaks on a small pool, it will never survive at scale. 2️⃣ 𝗧𝗲𝘀𝘁 𝗥𝗼𝗯𝘂𝘀𝘁𝗻𝗲𝘀𝘀 (𝗛𝘆𝗯𝗿𝗶𝗱𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵𝗲𝗿): Can a recipe built for SWE transfer to deep research? We trained a compact agent to bridge structured enterprise tables (SQL) and unstructured web search in a single reasoning chain, showing the recipe generalizes across domains. 🛡️ 𝗖𝘂𝗿𝗮𝘁𝗲 𝗗𝗮𝘁𝗮 & 𝗞𝗶𝗹𝗹 𝗥𝗲𝘄𝗮𝗿𝗱 𝗛𝗮𝗰𝗸𝗶𝗻𝗴: Across both SWE and deep research, agents quickly learn to cheat—whether by digging through `git log` for gold patches or answering from memory to skip SQL queries. Strict environment isolation, raw format gating, and high-signal data curation beat raw data volume every time. 3️⃣ 𝗦𝗰𝗮𝗹𝗲 𝗗𝗲𝗻𝘀𝗲 𝘁𝗼 𝗠𝗼𝗘 (𝗙𝗹𝗮𝘀𝗵𝗡𝗲𝘅𝘁-𝗥𝗟): Once the recipe is rock-solid, we scale. Our latest paper details how we scale agentic RL from compact dense models to massive Mixture-of-Experts (MoE) architectures through optimized asynchronous training infrastructure. Agentic RL training is a core focus for the Snowflake AI Research team. If you're passionate about pushing the boundaries of RL modeling, MoE architectures, and agent systems—𝗰𝗼𝗺𝗲 𝗷𝗼𝗶𝗻 𝘂𝘀! 👇 - Arctic-SWE: https://www.snowflake.com/en/blog/engineering/arctic-swe-agentic-rl/ - HybridResearcher: https://www.snowflake.com/en/blog/engineering/hybridresearcher-deep-research-agents/ - paper https://arxiv.org/abs/2610.28998 - jobs https://careers.snowflake.com/us/en/job/SNCOUS307EB823D34D4A99B277A616163A4107EXTERNALENUSDEAEBF034681417BA870648D0D6A2A94/$%7Btitle%7D