ReVision: New Method Reduces Visual Token Usage in Computer-Use Agents by 46%
Researchers introduced ReVision, a technique that removes redundant visual information from screenshots to reduce token usage in computer-use agents by 46% while improving performance by 3%. Computer-use agents process graphical interfaces through visual tokens, which accumulate rapidly as interaction histories grow, limiting context windows. This efficiency gain allows agents to incorporate longer interaction histories within fixed computational budgets.
ReVision addresses a fundamental inefficiency in computer-use agents (CUAs)—systems that interact with graphical user interfaces by processing screenshots. The method uses a learned patch selector to identify and remove redundant visual patches across consecutive screenshots while preserving spatial structure needed for model performance. Testing across three benchmarks (OSWorld, WebTailBench, and AgentNetBench) with the Qwen2.5-VL-7B model showed that processing 5-screenshot trajectories achieved 46% average token reduction and 3% success rate improvement compared to baselines without redundancy removal. The efficiency gains enable agents to process longer interaction histories within fixed context and compute budgets, addressing a previous limitation where incorporating historical observations provided minimal performance benefits. The research demonstrates that when visual redundancy is removed, agent performance continues to improve as more past observations are incorporated.
What's missing
The paper does not discuss potential limitations of the patch selector approach, such as failure modes when removing patches that appear redundant but contain subtle but important visual changes, or computational overhead of the redundancy detection mechanism itself. Additionally, generalization to other multimodal models beyond Qwen2.5-VL-7B and applicability to real-world deployment scenarios are not addressed.
What different sources said
- arXiv cs.CLCenter
Co-Evolving Skill Generation and Policy Optimization
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.