Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.11221 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911207223459840 |
|---|---|
| author | Li, Tao Hu, Jinlong Wang, Yang Liu, Junfeng Liu, Xuejun |
| author_facet | Li, Tao Hu, Jinlong Wang, Yang Liu, Junfeng Liu, Xuejun |
| contents | LLM-brained web agents offer powerful capabilities for web automation but face a critical cost-performance trade-off. The challenge is amplified by web agents' inherently complex prompts that include goals, action histories, and environmental states, leading to degraded LLM ensemble performance. To address this, we introduce WebRouter, a novel query-specific router trained from an information-theoretic perspective. Our core contribution is a cost-aware Variational Information Bottleneck (ca-VIB) objective, which learns a compressed representation of the input prompt while explicitly penalizing the expected operational cost. Experiments on five real-world websites from the WebVoyager benchmark show that WebRouter reduces operational costs by a striking 87.8\% compared to a GPT-4o baseline, while incurring only a 3.8\% accuracy drop. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_11221 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | WebRouter: Query-specific Router via Variational Information Bottleneck for Cost-sensitive Web Agent Li, Tao Hu, Jinlong Wang, Yang Liu, Junfeng Liu, Xuejun Computation and Language LLM-brained web agents offer powerful capabilities for web automation but face a critical cost-performance trade-off. The challenge is amplified by web agents' inherently complex prompts that include goals, action histories, and environmental states, leading to degraded LLM ensemble performance. To address this, we introduce WebRouter, a novel query-specific router trained from an information-theoretic perspective. Our core contribution is a cost-aware Variational Information Bottleneck (ca-VIB) objective, which learns a compressed representation of the input prompt while explicitly penalizing the expected operational cost. Experiments on five real-world websites from the WebVoyager benchmark show that WebRouter reduces operational costs by a striking 87.8\% compared to a GPT-4o baseline, while incurring only a 3.8\% accuracy drop. |
| title | WebRouter: Query-specific Router via Variational Information Bottleneck for Cost-sensitive Web Agent |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2510.11221 |