Metraipolis › archive › dc_d70cc68224bd

DeciCAD: Harness-Grounded Learning of Action Policies for CAD Agents — arXiv:2610.21755

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.21755
License (as archived)
CC0-1.0
Archived
2026-10-16T09:03:56.931Z
SHA-256 of the archived text
d99172602f370cb3feb585cb5ffa0d9d7372516637e56cec8d281218ff09b009
Stable URL of this page
https://metraipolis.com/documento/dc_d70cc68224bd

How to cite

DeciCAD: Harness-Grounded Learning of Action Policies for CAD Agents — arXiv:2610.21755 — arXiv (en), https://arxiv.org/abs/2610.21755 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d70cc68224bd (Metraipolis archive dc_d70cc68224bd, sha256 d99172602f370cb3feb585cb5ffa0d9d7372516637e56cec8d281218ff09b009)
BibTeX
@misc{dcd70cc68224bd,
  title = {DeciCAD: Harness-Grounded Learning of Action Policies for CAD Agents — arXiv:2610.21755},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.21755},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d70cc68224bd; sha256 d99172602f370cb3feb585cb5ffa0d9d7372516637e56cec8d281218ff09b009},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.21755. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large Language Model agents operate through closed-loop action policies whose decision space is defined by an agent harness. We investigate how this operating space shapes agent behavior and policy learning in computer-aided design (CAD), a professional domain characterized by persistent state, geometric constraints, and long-horizon procedures. We instantiate this perspective in DeciCAD, a domain-native harness that exposes semantic CAD actions, decision-oriented observations, environment feedback, and persistent state transitions. Building on this harness, we develop a learning pipeline that extracts causal decision supervision from interaction trajectories and improves the policy through supervised fine-tuning (SFT) and geometry-grounded online reinforcement learning (RL). Across four CAD benchmarks, DeciCAD achieves the best F1 and intersection over union (IoU) while reducing model-output tokens by 45--88% relative to a general-purpose coding harness. Under matched supervision, its SFT policy improves F1 from 0.6883 to 0.7437 and outperforms the coding policy; subsequent online RL further raises geometry availability from 66.5% to 73.5% and F1 to 0.7880. These results demonstrate that the decision space induced by a harness materially affects both agent execution and policy learning in CAD.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Machine Learning (cs.LG); Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2610.21755 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T08:44:17.786Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).