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

This paper introduces QuoteBench, a benchmark designed to expose the gap between matched execution scores and true task success for LLM coding agents that issue Bash commands through serializing interfaces. Across 56 one-shot tasks from 14 incident-derived families, the authors cross the generation contract with the execution transport around one deliberately unescaped added parser. Replaying the same reply through this parser lowers success 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, but zero or slightly negative for two others. Raw generation is nearly saturated at the frontier, so boundary adaptation is what still separates models. For example, GPT-5.6-sol shows a matched gap of only -3.6 points yet hides -64.3 points of damage and +60.7 points of compensation. The authors conclude that deployment configurations reorder model rankings, urging evaluators to report model configuration, generation contract, execution path, operating point, and final-state validator rather than treating matched scores as intrinsic model properties.

QuoteBench: How Matched Scores Can Hide Command-Path Failures

Field: Machine Learning

Authors: Shangao Li, Yao Zhang, Volker Tresp, Yuanyuan Yang

Date: 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.

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.

Key Findings

  • Matched scores are misleading. Surface-level execution-match metrics can hide large underlying command-path failures introduced by serialization, wrapping, and reparsing.
  • QuoteBench isolates the boundary. It uses 56 one-shot tasks from 14 incident-derived families and exact final-state validation, crossing the generation contract with the execution transport around one deliberately unescaped added parser.
  • Replay damage is large. Running the same reply through the added parser cuts success by 55.4–73.2 percentage points across eight same-window configurations.
  • Disclosure recovery varies. Telling the model about the boundary recovers 30.4–60.7 points for six configurations, but zero or slightly negative for the remaining two — adaptation, not raw generation, differentiates frontier models.
  • A concrete example. GPT-5.6-sol shows only a -3.6-point matched gap, which conceals -64.3 points of damage and +60.7 points of compensation.
  • Rankings depend on deployment. Among 26 comparable model pairs, one ranking reversal is unambiguous and four more sit on single-task margins, so deployment configuration changes leaderboards.
  • Evaluation recommendations. Command-issuing agent evaluations should report the model configuration, generation contract, execution path, operating point, and final-state validator — not 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 2026-08-15*

Tags

#arxiv#machine-learning#llm-agents#benchmark#bash-commands#evaluation#command-parsing

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