Metraipolis › archive › dc_da88ba3ff81d
Does an Agent's History Tell You When Compaction Will Hurt? A Modest, Bounded Effect on the TRACE Paired-Replay Corpus — arXiv:2610.08722
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.08722
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T11:19:54.293Z
- SHA-256 of the archived text
49dff24cf7675465d8fe3477f68bc995bfe9f531016e05786cc06992b5dabff3- Stable URL of this page
https://metraipolis.com/documento/dc_da88ba3ff81d
How to cite
Does an Agent's History Tell You When Compaction Will Hurt? A Modest, Bounded Effect on the TRACE Paired-Replay Corpus — arXiv:2610.08722 — arXiv (en), https://arxiv.org/abs/2610.08722 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_da88ba3ff81d (Metraipolis archive dc_da88ba3ff81d, sha256 49dff24cf7675465d8fe3477f68bc995bfe9f531016e05786cc06992b5dabff3)
BibTeX
@misc{dcda88ba3ff81d,
title = {Does an Agent's History Tell You When Compaction Will Hurt? A Modest, Bounded Effect on the TRACE Paired-Replay Corpus — arXiv:2610.08722},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.08722},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_da88ba3ff81d; sha256 49dff24cf7675465d8fe3477f68bc995bfe9f531016e05786cc06992b5dabff3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.08722. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Many long-horizon agents compact their context on a global rule, usually a token budget, blind to what the agent was doing. We ask whether the agent's recent behaviour predicts when a compaction will hurt. TRACE's public corpus of 590 harness-triggered AppWorld compaction boundaries replays each boundary from a re-executed prefix state under the pre-compaction context and under the summary, and records the burden of the next actions: calls that error or repeat a call already made. We find that pre-boundary history predicts post-compaction harm only weakly. An internally prespecified contrast by prefix placement is a wide null, and the naive "has-written" label behind it turns out to measure trajectory phase. The best extension-protocol trigger reaches held-out AUROC 0.66 (0.64 on the replicate's own label) against a same-boundary replicate of 0.72; the best frozen, interpretable trigger avoids 21% of harmful (positive-burden) boundaries while keeping 84% of compaction opportunities, and exceeds the random-rule expectation on count but not on burden mass (a post hoc comparison). Whether the best trigger beats a token-budget rule at matched retention cannot be evaluated on the release. We state what corpora should ship to answer it. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.08722 — 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 2026-10-07T11:15:24.632Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).