Beyond Outcomes: Dual-View Relational Learning for Efficient Agent Benchmarking
[Przetwarzanie języka] Autorzy: Xinshuai Guo, Junjie Wu, Dolly Deng i 4 więcej
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 $24\times$--$40\times$ compressio…
- Kategoria arxiv
- cs.CL
- Kategoria
- Przetwarzanie języka