HiViG is a test-time critic that combines macro-action history summarization with visual grounding of execution coordinates to reduce short-sighted and visually erroneous actions in long-horizon GUI agents.
Gui-shepherd: Reliable process reward and verification for long-sequence gui tasks, September 2025
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
other 1
citation-polarity summary
years
2026 6verdicts
UNVERDICTED 6roles
other 1polarities
unclear 1representative citing papers
BBCritic reframes GUI critique as continuous semantic alignment via contrastive learning in an affordance space, outperforming larger binary SOTA models on a new four-level hierarchical benchmark without extra annotations.
EVPO adaptively switches between critic-based and batch-mean advantage estimation using batch-level explained variance to provably achieve no greater variance than the better of PPO or GRPO at every step.
VisCritic uses visual comparison of pre- and post-action GUI screenshots via a Siamese vision transformer and Action-Aware Critic Head to provide process rewards, improving agent performance on benchmarks.
ChainWorld builds 347 chains from atomic OSWorld tasks and benchmarks four agents under single-turn and multi-turn protocols, reporting a maximum 31% completion rate with distinct failure profiles.
StainFlow proposes global entity stain tracking and local stain evidence linking modules to improve process rewards for GUI agents, reporting 3.2% relative gain in online RL success and 1.8% in judgment accuracy on AndroidWorld and OGRBench.
citing papers explorer
-
A History-Aware Visually Grounded Critic for Computer Use Agents
HiViG is a test-time critic that combines macro-action history summarization with visual grounding of execution coordinates to reduce short-sighted and visually erroneous actions in long-horizon GUI agents.
-
Beyond Binary: Reframing GUI Critique as Continuous Semantic Alignment
BBCritic reframes GUI critique as continuous semantic alignment via contrastive learning in an affordance space, outperforming larger binary SOTA models on a new four-level hierarchical benchmark without extra annotations.
-
EVPO: Explained Variance Policy Optimization for Adaptive Critic Utilization in LLM Post-Training
EVPO adaptively switches between critic-based and batch-mean advantage estimation using batch-level explained variance to provably achieve no greater variance than the better of PPO or GRPO at every step.
-
VisCritic: Visual State Comparison as Process Reward for GUI Agents
VisCritic uses visual comparison of pre- and post-action GUI screenshots via a Siamese vision transformer and Action-Aware Critic Head to provide process rewards, improving agent performance on benchmarks.
-
ChainWorld: Composing Long-Horizon Desktop Workloads from Atomic OSWorld Tasks
ChainWorld builds 347 chains from atomic OSWorld tasks and benchmarks four agents under single-turn and multi-turn protocols, reporting a maximum 31% completion rate with distinct failure profiles.
-
StainFlow: Entity-Stain Tracking and Evidence Linking for Process Rewards in GUI Agents
StainFlow proposes global entity stain tracking and local stain evidence linking modules to improve process rewards for GUI agents, reporting 3.2% relative gain in online RL success and 1.8% in judgment accuracy on AndroidWorld and OGRBench.