| title | Mid-Harness: Scaling Actions Between Model and Harness for Terminal Agents |
|---|---|
| authors | Minki Kang, Ryo Hachiuma, Shaokun Zhang, Subhashree Radhakrishnan, Yonggan Fu, Jindong Jiang, Mingjie Liu, Ehsan Hosseini-Asl, Yi Dong, Yu-Chiang Frank Wang, Byung-Kwan Lee |
| date | 2026-09-30 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2609.39982 |
| affiliation | NVIDIA |
| upvotes | 95 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf |
| abstract | Terminal agents act through stochastic model generations, yet the ability to generate a useful action does not ensure its reliable execution. A poor command (e.g., wrong package install) can change the environment in ways that hinder subsequent progress, even when the model could generate a better alternative. We investigate whether allocating test-time compute at the model-harness boundary can improve action reliability and trajectory success, and what makes this allocation effective. To study these questions, we introduce Mid-Harness, which samples and verifies candidate actions before forwarding one for execution, while keeping the generator and harness unchanged. With a TMAX-9B generator, more action sampling yields little benefit under weak verification, whereas a capable verifier can exploit useful alternatives from the same generator. On TerminalBench-Lite, a GPT-5.6 Sol verifier raises Pass@1 from 50.00% for the base agent to 68.03% with 8 sampled actions. When the same TMAX-9B model serves as the verifier, pairwise verification performs best among the evaluated verification mechanisms. Distilling responses from the stronger verifier into TMAX-9B further improves Pass@1, while leaving the action generator unchanged. With TMAX-9B on TerminalBench-Lite, combining action and trajectory scaling reaches higher success at lower estimated token cost than generating more trajectories alone. Mid-Harness also improves performance across additional models, benchmarks, and harnesses. These findings identify action scaling as a promising target for test-time compute scaling in terminal agents. |
cite
@misc{kang2026mid,
title={Mid-Harness: Scaling Actions Between Model and Harness for Terminal Agents},
author={Minki Kang and Ryo Hachiuma and Shaokun Zhang and Subhashree Radhakrishnan and Yonggan Fu and Jindong Jiang and Mingjie Liu and Ehsan Hosseini-Asl and Yi Dong and Yu-Chiang Frank Wang and Byung-Kwan Lee},
year={2026},
eprint={2609.39982},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2609.39982}
}