Pith. sign in

REVIEW 7 cited by

Large Language Models Can Self-Improve At Web Agent Tasks

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2405.20309 v2 pith:WOXYTTAQ submitted 2024-05-30 cs.LG cs.AIcs.CL

classification cs.LGcs.AIcs.CL
keywords modelsagentagentsbenchmarkdatalanguagellmsnavigate
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Training models to act as agents that can effectively navigate and perform actions in a complex environment, such as a web browser, has typically been challenging due to lack of training data. Large language models (LLMs) have recently demonstrated some capability to navigate novel environments as agents in a zero-shot or few-shot fashion, purely guided by natural language instructions as prompts. Recent research has also demonstrated LLMs have the capability to exceed their base performance through self-improvement, i.e. fine-tuning on data generated by the model itself. In this work, we explore the extent to which LLMs can self-improve their performance as agents in long-horizon tasks in a complex environment using the WebArena benchmark. In WebArena, an agent must autonomously navigate and perform actions on web pages to achieve a specified objective. We explore fine-tuning on three distinct synthetic training data mixtures and achieve a 31\% improvement in task completion rate over the base model on the WebArena benchmark through a self-improvement procedure. We additionally contribute novel evaluation metrics for assessing the performance, robustness, capabilities, and quality of trajectories of our fine-tuned agent models to a greater degree than simple, aggregate-level benchmark scores currently used to measure self-improvement.

Discussion (0). Sign in to comment.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Skill Self-Play: Pushing the Frontier of LLM Capability with Co-Evolving Skills

    cs.CL 2026-07 conditional novelty 6.0 of 10

    Skill Self-Play uses an evolving skill library to guide LLM self-training, improving tool-call and reasoning accuracy beyond unguided self-play on five model backbones.

  2. Morae: Proactively Pausing UI Agents for User Choices

    cs.HC 2025-08 conditional novelty 6.0 of 10

    Morae, a UI agent that proactively pauses at ambiguous decision points, helps blind and low-vision users complete more tasks and express preferences better than fully autonomous agents.

  3. Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Application

    cs.CL 2026-06 unverdicted novelty 5.0 of 10

    This survey categorizes agentic environments for LLMs by eight attributes and domains, introduces symbolic and neural synthesis paradigms with evaluation, and outlines four agent evolution pathways plus three environm...

  4. Rethinking Token Pruning for Historical Screenshots in GUI Visual Agents: Semantic, Spatial, and Temporal Perspectives

    cs.CV 2026-03 unverdicted novelty 5.0 of 10

    Empirical study finds background semantics, random pruning, and recency-based allocation improve token efficiency for GUI visual agents.

  5. Towards Understanding, Analyzing, and Optimizing Agentic AI Execution: A CPU-Centric Perspective

    cs.AI 2025-11 conditional novelty 5.0 of 10

    The paper analyzes CPU bottlenecks in agentic AI serving, selects representative workloads, and demonstrates that CPU-aware scheduling optimizations COMB and MAS can reduce P50 latency by up to 1.7x and total latency ...

  6. Plan-and-Act: Improving Planning of Agents for Long-Horizon Tasks

    cs.CL 2025-03 unverdicted novelty 5.0 of 10

    Plan-and-Act trains a dedicated Planner on synthetic plan-annotated trajectories to generate high-level plans that an Executor follows, reaching 57.58% success on WebArena-Lite and 81.36% on WebVoyager.

  7. Compute-Accuracy Pareto Frontiers for Open-Source Reasoning Large Language Models

    cs.CL 2025-12 conditional novelty 4.0 of 10

    Across five reasoning benchmarks, sparse MoE models dominate the accuracy-vs-FLOPs Pareto frontier, inference-compute gains saturate, and wrong answers systematically consume more compute than correct ones.

Pith tools