← 研踪 PaperTrailarXiv:2610.15884 · v4
RetryBench: Separating Retry Permission from Recovery Outcomes in GPT Tool Use
首次提交 2026-10-12 04:00
版本更新 2026-11-04 15:00
雷达观测 2026-11-05 11:10
摘要
LLM agents that call external tools must decide what to do when a tool call times out: retrying risks duplicate side effects, while not retrying can leave the task incomplete. We present RetryBench, an evaluation protocol for retry decisions and failure recovery in tool-using LLM agents. Each decision is scored on retry permission (is the retry justified by an idempotency-key guarantee or proof of non-execution?), settled effects (duplicates and omissions), and completion, by replaying it across all hidden execution worlds consistent with the observation. On four synthetic software-engineering tasks, we run two controlled prompt-sensitivity studies (1,728 decisions). Study 1 (idempotency-key descriptions): with the contract fixed, GPT-5.6 Luna reuses the valid idempotency key in 10/40 decisions when an extra key is described as a "Candidate recovery key" but in 39/40 when it is an "Alternative key string"; across three further phrasings per family the gap is 27/120 versus 113/120. GPT-5.6 Sol and Terra are at ceiling (40/40). Study 2 (action order): under underspecified lookup guarantees, retry rates are 17-27 points lower for GPT-5.6 Terra and GPT-5.4 when retry is listed first (option-order bias), mainly on two tasks; we make no claim about which behavior is better. Findings are limited to researcher-chosen phrasings, four tasks and unpinned model aliases. Code and all model responses are released.
作者备注：LLM agents, tool use, idempotency, retry, failure recovery, evaluation, prompt sensitivity, position bias, reliability
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v4 · 2026-11-04 15:00 · 最新观测版本
RetryBench: Separating Retry Permission from Recovery Outcomes in GPT Tool Use
LLM agents that call external tools must decide what to do when a tool call times out: retrying risks duplicate side effects, while not retrying can leave the task incomplete. We present RetryBench, an evaluation protocol for retry decisions and failure recovery in tool-using LLM agents. Each decision is scored on retry permission (is the retry justified by an idempotency-key guarantee or proof of non-execution?), settled effects (duplicates and omissions), and completion, by replaying it across all hidden execution worlds consistent with the observation. On four synthetic software-engineering tasks, we run two controlled prompt-sensitivity studies (1,728 decisions). Study 1 (idempotency-key descriptions): with the contract fixed, GPT-5.6 Luna reuses the valid idempotency key in 10/40 decisions when an extra key is described as a "Candidate recovery key" but in 39/40 when it is an "Alternative key string"; across three further phrasings per family the gap is 27/120 versus 113/120. GPT-5.6 Sol and Terra are at ceiling (40/40). Study 2 (action order): under underspecified lookup guarantees, retry rates are 17-27 points lower for GPT-5.6 Terra and GPT-5.4 when retry is listed first (option-order bias), mainly on two tasks; we make no claim about which behavior is better. Findings are limited to researcher-chosen phrasings, four tasks and unpinned model aliases. Code and all model responses are released.
查看此版本解读 →查看此版本原文 ↗