腾讯发布环境进化论文解决智能体训练难题
Banger paper from Tencent on environment evolution. Environment supply is becoming the main limit o...
腾讯这篇论文解决了智能体RL的环境供应瓶颈，提出不依赖智能体弱点就能提升环境难度的方法。
腾讯提出环境进化方法解决智能体强化学习中的环境供应限制。该方法从多轮训练目标直接推导三种提升环境难度的方法，并在固定计划上逐代应用。在Qwen3.6-27B和Qwen3.6-35B-A3B模型上应用后，Terminal-Bench 2.1基准得分提升14.4和18.0点。Hy4预览版、Claude Opus 5和GPT-5.6 Sol在进化环境中表现更差，证实了环境难度提升。
Banger paper from Tencent on environment evolution. Environment supply is becoming the main limit o...
Banger paper from Tencent on environment evolution. Environment supply is becoming the main limit on agent RL. So this is worth a read. (bookmark it) Agent RL needs a steady supply of environments hard enough to teach something new. Recent methods build them from the weaknesses an agent shows during its own rollouts. That approach has two problems. The environments inherit the agent's blind spots, so they generalize poorly. And as the agent improves it exposes fewer weaknesses to build from, so the learning signal weakens. Environment evolution raises difficulty without watching the agent at all. The paper derives three ways to make an environment harder straight from the multi-turn training objective, then applies them generation by generation on a fixed schedule. They test the generator before trusting it. Hy4 preview, Claude Opus 5 and GPT-5.6 Sol all do worse on the evolved environments, which confirms the difficulty. Plain long-horizon RL on Qwen3.6-27B and Qwen3.6-35B-A3B then adds 14.4 and 18.0 points on Terminal-Bench 2.1. Paper: academy.dair.ai/papers/environ… 💬 1 🔄 0 ❤️ 3 👀 729 📊 1 ⚡
