The authors present WebMRE, an offline of 541 web tasks and 5,293 steps derived from successful WebArena trajectories. Each step pairs a human-oriented guide sentence with a grounded action. The benchmark uses audited labels and a deterministic scoring protocol, enabling reproducible comparisons without live environments. They study the mutual reinforcement effect between guide and action. Joint decoding of guide and action improves element selection over an action-only baseline: for Qwen3.5-4B, gains of 0.9 points (action-first) and 0.2 points (guide-first); for Qwen3.5-9B, gains of 1.7 and 2.2 points. The effect holds across seeds and grows with model scale. Mediation analysis shows the guide is a causal channel. Forcing the gold guide as a decoding prefix lifts action accuracy from .422 to .684; another step's guide collapses it to .055. A paraphrase that renames the target recovers half the gain, showing the guide carries instruction meaning. Fine-tuned models outperform GPT-5.5, Claude Opus 4.8, and Gemini 3.5 Flash on all offline metrics.
Generating user guides can improve web agents’ action selection
On a deterministic offline benchmark, jointly training models to produce instructions and grounded actions modestly improved element selection, especially at larger scale.
Research Lab
Chengguang Gan · Yunhao Liang · QingHao Zhang · Shiwen Ni
Independent Researcher · University of Chinese Academy of Sciences · Pusan National University · Shenzhen University of Advanced Technology
Research Digest··2 min read
In one line
Jointly decoding a guide and action improves web agent element selection, and guides causally drive actions.
Before you read this
Three ideas this paper builds onco-evolving feedbackFeedback that evolves alongside the policy or components it guides to improve performance.43 papers →benchmarkA standard set of tasks or data used to evaluate and compare the performance of AI systems.366 papers →vision-language-action modelA robot policy model that maps visual observations and language instructions to actions, trained on demonstrations and often used for manipulation tasks.87 papers →
Read it as
Gan et al. built WebMRE from successful WebArena trajectories to study web-agent training without the variability of live websites or model-based judges.
The authors assembled 541 tasks containing 5,293 steps from successful WebArena trajectories.
How it works
before and afterThe Daily Edition
The front page in your inbox each weekday morning: the lead, the picks, and why each one earned its place.