Beyond Outcomes: Dual-View Relational Learning for Efficient Agent Benchmarking

2026-09-16

AI Summary

本文提出DualViewEval方法，通过联合利用结果和过程关系来压缩代理基准测试，有效减少了评估成本并提高了准确性。

Abstract

Agent benchmarks are substantially more costly to evaluate than conventional LLM benchmarks. Benchmark compression is therefore a natural solution, yet existing methods primarily model redundancy in task--model final-score distributions, which is important in agentic evaluation. To address this limitation, we analyze large-scale trajectories and identify six complementary process signals that are systematically associated with final agent performance. To disentangle agent performance redundancy from a complete perspective, we propose DualViewEval, an agent benchmark compression method that jointly exploits outcome and process relations to learn an exact-size miniset and predict the full-benchmark scores. Across five agent benchmarks and five representative baselines, DualViewEval achieves the best results in all datasets. With only 20 tasks, it achieves $24\times$--$40\times$ compression on APEX-Agents and BFCL, reducing mean absolute error (MAE) by $14.5\%$--$28.2\%$ over the strongest competitors while improving Kendall's $τ$ by up to $7.2\%$ relative to EssenceBench on SWE-bench Verified. The selected minisets further reveal capability differences among different agents, providing compact and diagnostic feedback for efficient agentic model development.

Problem

• agent benchmarking
• benchmark compression
• final-score distributions
• performance redundancy
• trajectory analysis

Innovation

• DualViewEval
• agent benchmark compression
• outcome and process relations
• miniset learning