AI Paper Atlas

Research map / Multi-agent systems and games / LLM agents

WebRetriever: A Large-Scale Comprehensive Benchmark for Efficient Web Agent Evaluation

Wei Dong, Tianyu Fu, Zhe Yu, Hanning Wang, Anyang Su, Zhizhou Fang, Yuyang Chen, Shuo Wang, Minghui Wu, Ping Jiang, Zhen Lei, Chenxu Zhao

ECCV 2026 · Multi-agent systems and games › LLM agents · approach: Agentic benchmarks and orchestration · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add operational documentation (Gemini-2.5-Pro) (ablation)Protocol I agent · WebRetrieversuccess rate (%): 40.9 -> 49.2Improved
add operational documentation (Claude-4.5) (ablation)Protocol I agent · WebRetrieversuccess rate (%): 31.3 -> 39.7Improved
add operational documentation (Gemini-2.5-Pro) (ablation)Protocol II agent · WebRetrieversuccess rate (%): 41.4 -> 50.1Improved
add operational documentation (Claude-4.5) (ablation)Protocol II agent · WebRetrieversuccess rate (%): 31.9 -> 40.1Improved
require information extraction after navigation (Gemini-2.5-Pro) (ablation)Protocol III agent · WebRetrieversuccess rate (%): 43.0 -> 21.0Hurt
require information extraction after navigation (Claude-4.5) (ablation)Protocol III agent · WebRetrieversuccess rate (%): 34.0 -> 16.0Hurt
add rule-based filtering of network requests (ablation)NavEval · WebRetrieveraverage human agreement rate (%): 87.8 -> 91.2Improved
use Claude-4.5-Sonnet instead of GPT-4o as judge (ablation)NavEval · WebRetrieveraverage human agreement rate (%): 86.4 -> 91.5Improved
use Claude-4.5-Sonnet instead of O4-mini as judge (ablation)NavEval · WebRetrieveraverage human agreement rate (%): 89.0 -> 91.5Improved
add network request signals (NavEval vs WebJudge)LLM-as-judge · WebRetrieveraverage human agreement rate (%): 81.0 -> 91.2Improved
NavEval vs best baseline judge on Online-Mind2WebLLM-as-judge · Online-Mind2Webaverage human agreement rate (%): 87.2 -> 97.0Improved
move from navigation-only (Protocol I) to end-to-end task completion (Protocol III)web agents · WebRetrieveraverage success rate (%): 21.1 -> 11.8Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy