Metraipolis › archive › dc_ffa53122bdb3

AgentFold: Closed-Loop Agentic Search for Protein Folding Model Design — arXiv:2608.26747

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.26747
License (as archived)
CC0-1.0
Archived
2026-10-17T14:22:56.712Z
SHA-256 of the archived text
b8912565aee8365176bef1099f8aed22e3377141ded7d504b4e8f17ad30ba9b2
Stable URL of this page
https://metraipolis.com/documento/dc_ffa53122bdb3

How to cite

AgentFold: Closed-Loop Agentic Search for Protein Folding Model Design — arXiv:2608.26747 — arXiv (en), https://arxiv.org/abs/2608.26747 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ffa53122bdb3 (Metraipolis archive dc_ffa53122bdb3, sha256 b8912565aee8365176bef1099f8aed22e3377141ded7d504b4e8f17ad30ba9b2)
BibTeX
@misc{dcffa53122bdb3,
  title = {AgentFold: Closed-Loop Agentic Search for Protein Folding Model Design — arXiv:2608.26747},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.26747},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ffa53122bdb3; sha256 b8912565aee8365176bef1099f8aed22e3377141ded7d504b4e8f17ad30ba9b2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.26747. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Scientific LLM agents have shown promise in literature reasoning, tool use, and experiment planning, but it remains unclear whether they can autonomously improve large, tightly coupled scientific machine-learning systems through executable code changes and computationally expensive validation. We study this question in protein folding, where progress requires coordinated architectural modifications, multi-objective evaluation, and domain-aware interpretation. We present AgentFold, a multi-agent framework that formulates folding-model development as a closed-loop search over executable code variants. Starting from ESMFold, AgentFold proposes hypotheses, implements and debugs code-level modifications, evaluates model variants, analyzes experimental outcomes, and stores both successful and failed interventions in structured memory. An MCTS-style policy allocates computational resources across high-scoring search branches. On an engineering-scale protein-folding codebase comprising more than 2,000 lines of code, AgentFold explores approximately 80 model variants using approximately 5,000 GPU-hours and 170 million LLM tokens. Under a matched computational budget, AgentFold improves the best lDDT by 7.5% over independent Codex proposals and outperforms a random-search control. Beyond model improvement, the resulting intervention traces reveal recurring empirical design patterns: stable gains tend to arise from early, soft, learnable priors and gated refinement, whereas direct geometric perturbations and geometry-conditioned feedback often destabilize training. The code and experimental resources are publicly available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-28 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.26747 — 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 2026-10-17T13:16:53.515Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).