# Preprint measures coding repair failures caused by context compaction

_Published Monday, October 26, 2026 at 7:13 AM EDT · AI, Science · Latest · Tier 2 — Notable_

Replacing coding agents' full histories with summaries damaged 5-64% of 22 successful repairs across 14 summarizer backbones in the Compact-Bench preprint. Context compaction lets agents continue beyond their context windows, but can erase information needed to finish a task.

The benchmark compares full-history and summary-only continuations from the same repository state. It covers 35 SWE-bench repairs, 146 agentic sessions, and 106 production chains. Language-model summaries retained at most 16.0% of recurring details, compared with 79.1% for mechanical extraction. However, retaining details did not ensure successful continuation: with Kimi-K3 summarizing, the Kimi Code prompt preserved the most task continuations, while Claude Code retained the most recurring details.

## Sources

- [cs.AI updates on arXiv.org](https://arxiv.org/abs/2610.31710)

---
Canonical: https://techandbusiness.org/newswire/BvkJcWqFndaewwJ59Un60_
Published: 2026-10-26T11:13:58.258Z
Story chronology: 2026-10-26T04:00:00.000Z
Retrieved: 2026-10-26T13:41:41.225Z
Publisher: Tech & Business (techandbusiness.org)