paper
A History-Aware Visually Grounded Critic for Computer Use Agents
paperactiveprovisional
a-history-aware-visually-grounded-critic-for-computer-use-agents-0f9a6c09·1 events·first seen 7d agoAliases: A History-Aware Visually Grounded Critic for Computer Use Agents
Co-occurring entities
More like this (12)
GUI AgentsComputer-Using AgentLearning Red Agent Policy from Observations for Neurosymbolic Autonomous Cyber AgentsSemantic AgentHow AI Agents Reshape Knowledge Work: Autonomy, Efficiency, and Scopetool-augmented language agentsEvaluation Cards: An Interpretive Layer for AI Evaluation ReportingMulti-Turn Evaluation of Deep Research Agents Under Process-Level FeedbackLabVLA: Grounding Vision-Language-Action Models in Scientific LaboratoriesConcrete Problems in AI SafetyTowards a Science of AI Agent Reliabilitycomputer-use agents
Recent events (1)
HiViG: History-aware visually grounded critic improves computer use agents across GUI benchmarks
Researchers introduce HiViG, a test-time framework for Computer Use Agents that addresses two weaknesses in existing critic models: short-sighted decision loops and lack of visual grounding. The system trains a multimodal critic on real GUI trajectories to maintain a compact macro-action history and verify execution coordinates against live screenshots before action execution. Evaluated on web, mobile, and desktop benchmarks, HiViG improves average success rates by 5.8% over the strongest baseline with Qwen3-VL-32B and 9.0% with Gemini-3-Flash, with both history and grounding components shown to be independently necessary.