Meta-Harness v1b: Policy x Harness Co-Evolution
Collection
Co-evolution GRPO (verl, Qwen3.5): shared policy = grader/solver + harness-editor. IMO/paper-review/math. bf16 checkpoints + evolved scaffolds. • 20 items • Updated
Dataset Viewer
The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.
Meta-Harness v1b — Evolved Scaffolds
Harnesses (scaffolds) evolved by the Proposer during co-evolution, plus reconstructed top-K per domain.
coevo_scaffolds/— harnesses promoted during training (best-of-N per step).eval_*_topk/— top-K scaffolds reconstructed from rollout dumps, ranked by proposer accuracy. See the model Collection: https://huggingface.co/collections/hyunseoki/meta-harness-v1b-policy-x-harness-co-evolution-6a5c9877f74920e8922808bc
- Downloads last month
- 190