← 研踪 PaperTrailarXiv:2610.30499 · v2
Can LLMs Improve Their Own Harness? A Controlled Study on ARC-AGI-3
首次提交 2026-10-22 06:29
版本更新 2026-11-14 00:24
雷达观测 2026-11-16 11:41
摘要
A generally intelligent system acquires new skills efficiently without requiring humans to redesign its software scaffold for every new task. Yet language models rarely act alone. A harness is a symbolic program that orchestrates calls to the model, handling perception, memory, tool use, recovery, and inference allocation on the model's behalf. ARC-AGI-3 is an interactive benchmark in which agents must explore unfamiliar environments, figure out their goals on the fly, and keep learning as they go. On it, human-steered systems already report ceiling scores across the public games, so we know effective designs exist. We ask whether frontier models can build and improve such a harness on their own. Through a fixed coding bootstrap, each model constructs a complete task harness, acts through that harness on five development games, and revises the entire codebase twice. Every version is evaluated on twenty games that never influence adaptation. We measure task performance with Relative Human Action Efficiency (RHAE) and record dollar-denominated inference cost. Across 27 lineages, harness performance ranges from zero to near-ceiling scores. Two revisions raise mean held-out RHAE by only 2.80 points (95% interval [-1.44,7.35]). Development gains overstate held-out improvement, and reductions in inference cost can come at the expense of task performance. No revision retains its parent, yet a post hoc rule that selects among existing snapshots by development score raises the held-out gain to 4.53 points ([0.81,8.68]), separating the generation of revisions from the decision to deploy them. Human-steered systems demonstrate further headroom. These findings reveal a gap between autonomous harness construction and reliable adaptation under this two-revision protocol: even strong task performance does not ensure that subsequent revisions generalize or preserve existing capabilities.
作者备注:32 pages, 5 figures. Includes appendices and a supplementary code/data archive. v2: revised version with additional analyses
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-14 00:24 · 最新观测版本
Can LLMs Improve Their Own Harness? A Controlled Study on ARC-AGI-3
A generally intelligent system acquires new skills efficiently without requiring humans to redesign its software scaffold for every new task. Yet language models rarely act alone. A harness is a symbolic program that orchestrates calls to the model, handling perception, memory, tool use, recovery, and inference allocation on the model's behalf. ARC-AGI-3 is an interactive benchmark in which agents must explore unfamiliar environments, figure out their goals on the fly, and keep learning as they go. On it, human-steered systems already report ceiling scores across the public games, so we know effective designs exist. We ask whether frontier models can build and improve such a harness on their own. Through a fixed coding bootstrap, each model constructs a complete task harness, acts through that harness on five development games, and revises the entire codebase twice. Every version is evaluated on twenty games that never influence adaptation. We measure task performance with Relative Human Action Efficiency (RHAE) and record dollar-denominated inference cost. Across 27 lineages, harness performance ranges from zero to near-ceiling scores. Two revisions raise mean held-out RHAE by only 2.80 points (95% interval [-1.44,7.35]). Development gains overstate held-out improvement, and reductions in inference cost can come at the expense of task performance. No revision retains its parent, yet a post hoc rule that selects among existing snapshots by development score raises the held-out gain to 4.53 points ([0.81,8.68]), separating the generation of revisions from the decision to deploy them. Human-steered systems demonstrate further headroom. These findings reveal a gap between autonomous harness construction and reliable adaptation under this two-revision protocol: even strong task performance does not ensure that subsequent revisions generalize or preserve existing capabilities.
查看此版本解读 →查看此版本原文 ↗