Metraipolis › archive › dc_fa88bd9bb0e4
Absorbed in Inertia: Activation Analysis for Computer-Use Agents — arXiv:2609.37176
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.37176
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T16:23:35.415Z
- SHA-256 of the archived text
f6ff7850ff6de2b5d4a45e69d2dd8836ab9398c8336bedd509dcfd5e3d24f932- Stable URL of this page
https://metraipolis.com/documento/dc_fa88bd9bb0e4
How to cite
Absorbed in Inertia: Activation Analysis for Computer-Use Agents — arXiv:2609.37176 — arXiv (en), https://arxiv.org/abs/2609.37176 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fa88bd9bb0e4 (Metraipolis archive dc_fa88bd9bb0e4, sha256 f6ff7850ff6de2b5d4a45e69d2dd8836ab9398c8336bedd509dcfd5e3d24f932)
BibTeX
@misc{dcfa88bd9bb0e4,
title = {Absorbed in Inertia: Activation Analysis for Computer-Use Agents — arXiv:2609.37176},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.37176},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fa88bd9bb0e4; sha256 f6ff7850ff6de2b5d4a45e69d2dd8836ab9398c8336bedd509dcfd5e3d24f932},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.37176. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Computer-use agents have become increasingly capable of executing tasks on live desktops through natural-language instructions, based on trajectories of screenshots, actions, and reasoning. We discover that they can stealthily exhibit inertia, in which they repeat fruitless actions despite recognizing that these actions are ineffective. We hypothesize that inertia is reflected in the agent's internal state, i.e., the activation values of the agent's underlying model, and propose a protocol to measure the relationship between the two. Extensive analysis of high-dimensional activation states shows that inertia corresponds to an absorbing region of activation space, where activation values become stale across actions and even after attempts to steer them. We conjecture that drastically changing the agents' activations by re-initializing them is necessary to escape inertia. Specifically, we propose R$^3$ (Reset, Reroute, Restore), which temporarily resets the agent's context trajectory to escape the absorbing region and then restores the historical context to effectively complete the task. Our approach yields 17-55% lower measured inertia across models relative to unmodified agents. These results suggest that changing the context can interrupt recurrence more effectively than directly steering the resulting activations. Our code is available at this https URL Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.37176 — 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 2026-10-07T15:54:53.598Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).