Metraipolis › archive › dc_5fd3db031d09
RippleCP: Measuring Counterfactual Checkpoint Advantage in Coding Agents — arXiv:2610.09088
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.09088
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T11:08:54.318Z
- SHA-256 of the archived text
905de1b38a7a3fdff56a3702cb0650776424576913e928be239a6bd82f35f05b- Stable URL of this page
https://metraipolis.com/documento/dc_5fd3db031d09
How to cite
RippleCP: Measuring Counterfactual Checkpoint Advantage in Coding Agents — arXiv:2610.09088 — arXiv (en), https://arxiv.org/abs/2610.09088 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5fd3db031d09 (Metraipolis archive dc_5fd3db031d09, sha256 905de1b38a7a3fdff56a3702cb0650776424576913e928be239a6bd82f35f05b)
BibTeX
@misc{dc5fd3db031d09,
title = {RippleCP: Measuring Counterfactual Checkpoint Advantage in Coding Agents — arXiv:2610.09088},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.09088},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5fd3db031d09; sha256 905de1b38a7a3fdff56a3702cb0650776424576913e928be239a6bd82f35f05b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09088. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Agent checkpoint systems decide what state is recovery-relevant, how to snapshot it, and whether rollback is admissible. None decides which of the safe boundaries they expose are worth materializing. We formulate this as counterfactual checkpoint advantage, the reduction in future recovery cost obtained by checkpointing a candidate rather than skipping it, and measure it by driving a CP branch and a SKIP branch to the same logical failure and recovering both under matched model, tool, verifier, and stopping conditions. On a frozen pilot of 12 SWE-bench Verified tasks and 106 real recovery branches, checkpointing saves 49.4 s per task, and that figure resolves into two regimes two orders of magnitude apart. The first checkpoint returns 100.0 s on 156.5 s of protected work, a conversion of 0.64; a second one step later returns-1.1 s on 41.8 s, a conversion of -0.03. Recovery is a re-derivation rather than a replay, so preserved work is a poor guide to saved work, and the classical elapsed-work rule misprices the second checkpoint by its full nominal cost. We identify where placement can pay, and set the bar a placement policy must clear. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.09088 — 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 2026-10-08T10:14:05.899Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).