WALT: Web Agents that Learn Tools
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866912622681522176 |
|---|---|
| author | Prabhu, Viraj Dai, Yutong Fernandez, Matthew Gu, Jing Ramakrishnan, Krithika Luo, Yanqi Savarese, Silvio Xiong, Caiming Li, Junnan Chen, Zeyuan Xu, Ran |
| author_facet | Prabhu, Viraj Dai, Yutong Fernandez, Matthew Gu, Jing Ramakrishnan, Krithika Luo, Yanqi Savarese, Silvio Xiong, Caiming Li, Junnan Chen, Zeyuan Xu, Ran |
| contents | Web agents promise to automate complex browser tasks, but current methods remain brittle -- relying on step-by-step UI interactions and heavy LLM reasoning that break under dynamic layouts and long horizons. Humans, by contrast, exploit website-provided functionality through high-level operations like search, filter, and sort. We introduce WALT (Web Agents that Learn Tools), a framework that reverse-engineers latent website functionality into reusable invocable tools. Rather than hypothesizing ad-hoc skills, WALT exposes robust implementations of automations already designed into websites -- spanning discovery (search, filter, sort), communication (post, comment, upvote), and content management (create, edit, delete). Tools abstract away low-level execution: instead of reasoning about how to click and type, agents simply call search(query) or create(listing). This shifts the computational burden from fragile step-by-step reasoning to reliable tool invocation. On VisualWebArena and WebArena, WALT achieves higher success with fewer steps and less LLM-dependent reasoning, establishing a robust and generalizable paradigm for browser automation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_01524 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | WALT: Web Agents that Learn Tools Prabhu, Viraj Dai, Yutong Fernandez, Matthew Gu, Jing Ramakrishnan, Krithika Luo, Yanqi Savarese, Silvio Xiong, Caiming Li, Junnan Chen, Zeyuan Xu, Ran Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Web agents promise to automate complex browser tasks, but current methods remain brittle -- relying on step-by-step UI interactions and heavy LLM reasoning that break under dynamic layouts and long horizons. Humans, by contrast, exploit website-provided functionality through high-level operations like search, filter, and sort. We introduce WALT (Web Agents that Learn Tools), a framework that reverse-engineers latent website functionality into reusable invocable tools. Rather than hypothesizing ad-hoc skills, WALT exposes robust implementations of automations already designed into websites -- spanning discovery (search, filter, sort), communication (post, comment, upvote), and content management (create, edit, delete). Tools abstract away low-level execution: instead of reasoning about how to click and type, agents simply call search(query) or create(listing). This shifts the computational burden from fragile step-by-step reasoning to reliable tool invocation. On VisualWebArena and WebArena, WALT achieves higher success with fewer steps and less LLM-dependent reasoning, establishing a robust and generalizable paradigm for browser automation. |
| title | WALT: Web Agents that Learn Tools |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2510.01524 |