English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

QuoteBench: How Matched Scores Can Hide Command-Path Failures in LLM Coding Agents

Forum topic · 小凯 · 2026-08-15

Summary

QuoteBench is a benchmark that measures how execution-path failures are hidden behind matched execution scores in LLM coding agents that issue Bash commands through interfaces which may serialize, wrap, and reparse model output. Using exact final-state validation on 56 one-shot tasks from 14 incident-derived families, it crosses the generation contract with an execution transport containing a deliberately unescaped added parser. Replaying the same replies through the added parser lowers success rates by 55.4 to 73.2 percentage points across eight same-window configurations; disclosing the boundary recovers 30.4 to 60.7 points for six configurations, and zero or slightly negative for two others. Raw generation is nearly saturated at the frontier, so boundary adaptation is what still separates models. GPT-5.6-sol's matched gap of -3.6 points conceals -64.3 points of damage and +60.7 points of compensation. Deployment configuration reorders model rankings, and the authors argue evaluations should report configuration, generation contract, execution path, operating point, and validator rather than treat matched scores as intrinsic model properties. Paper: arXiv:2608.13547.

Paper Overview

Field: ML Authors: Shangao Li, Yao Zhang, Volker Tresp, Yuanyuan Yang Published: 2026-08-13 arXiv: 2608.13547

Summary

LLM coding agents issue Bash commands through interfaces that may serialize, wrap, and reparse model output. Matched execution scores alone cannot distinguish command-generation errors from failures introduced after generation. QuoteBench measures this boundary with exact final-state validation on 56 one-shot tasks from 14 incident-derived families, crossing the generation contract with the execution transport around one deliberately unescaped added parser.

Escaping at the interpolation point reproduces each replayed reply's raw-path outcome, so any recovery under a disclosed boundary must come from the model changing its generation.

Key Findings

  • Across eight same-window configurations, replaying the same reply through the added parser lowers success by 55.4 to 73.2 percentage points.
  • Disclosure of the boundary recovers 30.4 to 60.7 points for six configurations, and zero or slightly negative for the other two.
  • Raw generation is nearly saturated at the frontier; boundary adaptation is what still separates models.
  • GPT-5.6-sol's matched gap of -3.6 points hides -64.3 points of damage and +60.7 points of compensation.
  • The deployment configuration reorders models: one reversal among 26 comparable pairs is unambiguous, and four more sit on single-task margins.

Implication

Evaluations of command-issuing agents should report the model configuration, generation contract, execution path, operating point, and final-state validator rather than treat a matched score as an intrinsic model property.

Original Abstract

LLM coding agents issue Bash commands through interfaces that may serialize, wrap, and reparse model output. Matched execution scores alone cannot distinguish command-generation errors from failures introduced after generation. QuoteBench measures this boundary with exact final-state validation on 56 one-shot tasks from 14 incident-derived families, crossing the generation contract with the execution transport around one deliberately unescaped added parser. Escaping at the interpolation point reproduces each replayed reply's raw-path outcome, so any recovery under a disclosed boundary must come from the model changing its generation. Across eight same-window configurations, replaying the same reply through the added parser lowers success by 55.4 to 73.2 percentage points; disclosure recovers 30.4 to 60.7 points for six configurations, and zero or slightly negative for the other two. Raw generation is nearly saturated at the frontier; boundary adaptation is what still separates models. GPT-5.6-sol's matched gap of -3.6 points hides -64.3 points of damage and +60.7 points of compensation. The deployment configuration reorders models: one reversal among 26 comparable pairs is unambiguous and four more sit on single-task margins. Evaluations of command-issuing agents should report the model configuration, generation contract, execution path, operating point, and final-state validator rather than treat a matched score as an intrinsic model property.

--- *Auto-collected on 2026-08-15*

Tags

#quotebench#llm-coding-agents#benchmark#bash-commands#evaluation#arxiv#machine-learning#agent-evaluation

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178633500