papers / rollback-recovery

Rollback Recovery: Measuring Repair, Disclosure, and Restraint After Agent Errors

Pooria Arab · orcid 0009-0005-9964-6841 · 2026-10-29 · Published on Zenodo

abstract

Rollback Recovery measures whether an agent restores damaged state, reports the incident, and avoids needless repair. Its 48 synthetic episodes cover contact records, files, and messages, with 36 incidents and 12 clean controls. A visible-information scripted policy passes 48/48 tasks (100.0 [92.6, 100.0] percent; Wilson 95% interval). Removing structured reports leaves repair intact but reduces full success to 25.0 [14.9, 38.8] percent. Removing state tags reduces success to 45.8 [32.6, 59.7] percent. A simulated population sweep runs 24000 episodes and isolates report retention from repair ability. At half retention, simulated full success is 62.2 [60.9, 63.4] percent (bootstrap 95% interval across simulated agents). These are local scripted controls and simulated outcomes, not learned-model scores. Public templates, idealized reversibility, and unvalidated human interpretation limit generalization.

benchmark: benchmark-rollback-recovery. submitted to: none recorded.

citation

Arab, Pooria (2026). Rollback Recovery: Measuring Repair, Disclosure, and Restraint After Agent Errors. https://doi.org/10.5281/zenodo.23752421

bibtex

@misc{rollback-recovery,
  author = {Arab, Pooria},
  title = {Rollback Recovery: Measuring Repair, Disclosure, and Restraint After Agent Errors},
  year = {2026},
  doi = {10.5281/zenodo.23752421},
  url = {https://pooriaarab.com/papers/rollback-recovery},
}

venues, counts and the software records live on papers.