Tools & AgentsZhiyuan Fan et al.
Environment Evolution for Terminal Agents
Environment Evolution for Terminal Agents released — points out that existing co-evolution methods depend on on-policy rollouts, so the learning signal dries up as the model gets stronger, and presents a method that raises environment difficulty gradually off-policy and schedules environments evolved generation by generation
Why it matters
A companion paper to Terminal-Universe above, released the same week, splitting the problem of making environments (synthesis) from that of growing them (evolution). In the authors' own reporting, gains on Terminal-Bench 2.1 of 14.4 points for Qwen3.6-27B and 18.0 points for Qwen3.6-35B-A3B
Source — Primary sources first; secondary reports are labelled “Press”
Paperarxiv.orghttps://arxiv.org/abs/2609.04128Added in Week 37, 2026 · See that week