TRACE: Trajectory-robust Admission with Evidence Ordering for Efficient GUI Agents
TRACE, a training-free visual token pruning framework, cuts GUI agent inference latency and memory while keeping trajectory-wide visual evidence reusable.
TRACE is a training-free framework for trajectory-robust admission and coverage-aware evidence ordering that prunes high-resolution screenshot tokens accumulated in GUI agent trajectories. It ranks visual evidence using a query-independent layout-derived interaction prior combined with instruction relevance and feature novelty, and reserves part of the budget for native tokens distributed across the screen to repair spatial coverage. A monotone KV contraction incrementally compresses retired frames into compact session state, avoiding repeated visual encoding or pruning. Experiments across six GUI benchmarks and diverse models verify effectiveness under tight budgets, with source code to be released.
- Training-free visual token pruning for GUI agent screenshot trajectories
- Ranks evidence by layout prior, instruction relevance, and novelty
- Monotone KV contraction compresses retired frames without re-encoding
- Validated on six GUI benchmarks across diverse models
From the original article
1 min read at huggingface.co
GUI agents accumulate high-resolution screenshots as the trajectory unfolds, increasing inference latency and memory usage. Training-free visual token pruning can reduce this cost, but cache reuse introduces a fundamental constraint. Once tokens are discarded, the corresponding visual evidence cannot be recovered without re-encoding.
By Yuhao Wang, Mu Qiao, Xindong Zhang · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.