DRIVE: Modeling Skills at the Reasoning and Interaction Levels for Web Agents under Continual Learning

• URL: http://arxiv.org/abs/2605.23939v1
• Date: Tue, 28 Apr 2026 11:39:20 GMT
• Title: DRIVE: Modeling Skills at the Reasoning and Interaction Levels for Web Agents under Continual Learning
• Abstract summary: Web agents require both high-level reasoning and low-level interactions to conduct different tasks.<n>We propose DRIVE, a dual-level skill modeling framework separating historical experience into natural language reasoning skills.<n>Experiments show DRIVE attains an average task success rate of 52.8%, exceeding the skill-free baseline by 7.3 percentage points.
• Score: 17.92660876001036
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Web agents require both high-level reasoning (for task decomposition) and low-level interactions (for page elements manipulation) to conduct different tasks. However, these knowledge types differ fundamentally: reasoning knowledge (e.g., booking a flight requires first searching for routes) is abstract and transferable across websites, while interaction knowledge (e.g., clicking the Search button at a specific coordinate on Site A) depends heavily on page-specific contexts. Existing methods store experiences uniformly. This creates a dilemma: abstract representations lose executability on concrete pages, while concrete representations fail to generalize across domains. This entanglement limits capability accumulation: on new websites, agents either fail to recognize reusable task logic due to surface-level differences or attempt infeasible actions from outdated page structures. To disentangle them, we propose DRIVE, a dual-level skill modeling framework separating historical experience into natural language reasoning skills, which capture transferable task logic, and programmatic interaction skills, grounding abstract actions to executable operations. A scene-aware coordination mechanism adaptively retrieves and invokes these dual-level skills based on task semantics. DRIVE also uses skill-level reflection to identify hierarchy-specific failure modes, enabling targeted skill library expansion and refinement. Experiments across five WebArena domains show DRIVE attains an average task success rate of 52.8%, exceeding the skill-free baseline by 7.3 percentage points. Further ablations show reasoning and interaction skills provide distinct, complementary benefits, supporting separation of transferable task logic from executable page-level operations.

Related papers

• Beyond Domains: Reusing Web Skills via Transferable Interaction Patterns [68.0603867264595]
  SkillMigrator learns reusable web skills and transfers them across sites by matching layout structure rather than specific element references.<n>It reduces the average LLM-action count on successful trajectories by 8-10% across both WebArena and Mind2Web at matched success rate.
  arXiv  Detail & Related papers  (2026-06-16T08:04:35Z)
• Decompose and Recompose: Reasoning New Skills from Existing Abilities for Cross-Task Robotic Manipulation [45.91196295385836]
  Cross-task generalization is a core challenge in open-world robotic manipulation.<n>We propose Decompose and Recompose, a skill reasoning framework using atomic skill-action pairs as intermediate representations.<n>Our approach decomposes seen demonstrations into interpretable skill--action alignments, enabling the model to recompose these skills for unseen tasks.
  arXiv  Detail & Related papers  (2026-05-02T13:55:28Z)
• WebXSkill: Skill Learning for Autonomous Web Agents [104.76374637691212]
  WebXSkill is a framework that bridges the gap between code-based skills and natural language guidance.<n>On WebArena and WebVoyager, WebXSkill improves task success rate by up to 9.8 and 12.9 points over the baseline.
  arXiv  Detail & Related papers  (2026-04-14T21:48:15Z)
• Enhancing Web Agents with a Hierarchical Memory Tree [35.249218954574424]
  Large language model-based web agents have shown strong potential in automating web interactions through advanced reasoning and instruction following.<n>While retrieval-based memory derived from historical trajectories enables these agents to handle complex, long-horizon tasks, current methods struggle to generalize across unseen websites.<n>We identify that this challenge arises from the flat memory structures that entangle high-level task logic with site-specific action details.<n>We propose Hierarchical Memory Tree (HMT), a structured framework designed to explicitly decouple logical planning from action execution.
  arXiv  Detail & Related papers  (2026-03-07T04:07:47Z)
• SAME: Learning Generic Language-Guided Visual Navigation with State-Adaptive Mixture of Experts [54.11162991206203]
  This paper consolidates diverse navigation tasks into a unified and generic framework.<n>We propose a novel State-Adaptive Mixture of Experts (SAME) model that effectively enables an agent to infer decisions.
  arXiv  Detail & Related papers  (2024-12-07T06:12:53Z)
• DISCO: Embodied Navigation and Interaction via Differentiable Scene Semantics and Dual-level Control [53.80518003412016]
  Building a general-purpose intelligent home-assistant agent skilled in diverse tasks by human commands is a long-term blueprint of embodied AI research. We study primitive mobile manipulations for embodied agents, i.e. how to navigate and interact based on an instructed verb-noun pair. We propose DISCO, which features non-trivial advancements in contextualized scene modeling and efficient controls.
  arXiv  Detail & Related papers  (2024-07-20T05:39:28Z)
• SkillDiffuser: Interpretable Hierarchical Planning via Skill Abstractions in Diffusion-Based Task Execution [75.2573501625811]
  Diffusion models have demonstrated strong potential for robotic trajectory planning. generating coherent trajectories from high-level instructions remains challenging. We propose SkillDiffuser, an end-to-end hierarchical planning framework.
  arXiv  Detail & Related papers  (2023-12-18T18:16:52Z)
• Fast Inference and Transfer of Compositional Task Structures for Few-shot Task Generalization [101.72755769194677]
  We formulate it as a few-shot reinforcement learning problem where a task is characterized by a subtask graph. Our multi-task subtask graph inferencer (MTSGI) first infers the common high-level task structure in terms of the subtask graph from the training tasks. Our experiment results on 2D grid-world and complex web navigation domains show that the proposed method can learn and leverage the common underlying structure of the tasks for faster adaptation to the unseen tasks.
  arXiv  Detail & Related papers  (2022-05-25T10:44:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.