AutoEnvScaling turns reinforcement-learning environment design into a terminal task, letting a proposer agent build new environments from the solver's rollouts, raising the valid-environment rate by up to 25 times and cutting cost per valid environment by up to 75%.
Terminal-agent training data shifts from short command demonstrations to long-horizon trajectories averaging 52.8 turns, and fine-tuning Qwen3.5-27B on roughly 3K such trajectories raises Terminal-Bench 2.1 from 46.82 to 60.67 while beating the base model on all 12 benchmarks.
The work introduces CoTrace, a harness-aware data recipe for alternating model–harness co-evolution that explicitly governs trajectory routing, provenance matching, and curriculum refresh: execution failures are routed to harness synthesis while policy supervision uses only verified successes matched to the adopted runtime. On the 102-task Tmax promotion split, Qwen3.5-9B advances from 78 to 88 solved tasks under supervised fine-tuning and reaches 90 with an online reinforcement variant, whereas much larger corpora pooled across sibling harnesses yield no accepted model update.
The authors introduce software-in-the-loop reconstruction (SWR), a self-supervised framework that draws reference outputs and verification targets from existing software workflows, instantiated with 500 workflows across 46 software families and six domains, where Qwen3.8-Max solves 838 tasks over three attempts and yields 1,422 verified trajectories oversampled into 3,000 reconstruction-only training examples, and supervised fine-tuning raises Qwen3.8-27B mean Terminal-Bench 2 performance from 47.94% to 53.56% while achieving the highest mean among four matched-token corpus controls on all four reported evaluations.
The work introduces Mid-Harness, which samples and verifies candidate actions at the model-harness boundary before forwarding one for execution while keeping the generator and harness unchanged, and studies action-level test-time compute scaling: on TerminalBench-Lite a GPT-5.6 Sol verifier raises TMAX-9B Pass@1 from 50.00% to 68.03% with 8 sampled actions, wider sampling yields little benefit under weak verification, pairwise verification performs best among the evaluated self-verification mechanisms, distilling the stronger verifier's responses into TMAX-9B further improves Pass@1, and combining action scaling with trajectory scaling (Best-of-N and Sequential Refine) reaches higher success at lower estimated token cost.
You’re all caught up