Sudipta Paul, Pengrui Quan, Aounon Kumar, Nicholas Tzou, Vivek Kulkarni, Srinivas Chappidi · 2026 · arXiv
DOI: 10.48550/arXiv.2610.30042 · open-access PDF
Harness-Optimierung, die die Anrufe, Werkzeuge und Workflows eines Agenten verbessert, ist eine zunehmend effektive Möglichkeit, Agenten auf gefrorenen oder geschlossenen LLMs aufzubauen. Führende Ansätze wie die Darwin-Godel-Maschine und die Huxley-Godel-Maschine (HGM) verlassen sich auf recursive Selbstverbesserung und wurden hauptsächlich für das Codieren entwickelt, wo die erforderliche Kompetenz zur Verbesserung des Agenten mit der Verbesserung der Kompetenz übereinstimmt.
Machine translation (Meta M2M100); the English original follows.
Harness optimization, which improves an agent's prompts, tools, and workflows, is an increasingly effective way to enhance agents built on frozen or closed-source LLMs. Leading approaches such as the Darwin--Godel Machine and the Huxley--Godel Machine (HGM) rely on recursive self-improvement and were developed mainly for coding, where the competence needed to improve the agent coincides with the competence being improved. Planning tasks, in which an agent must produce a sequence of tool calls that jointly satisfies interacting constraints, break this alignment. There, open-ended self-modificat
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.30042, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange