On 108 long-horizon real-world computer workflows, frontier agents complete at most 20.6% of tasks and fail mainly by losing hidden state, not by basic GUI control.
hub Mixed citations
Aguvis: Unified Pure Vision Agents for Autonomous GUI Interaction
Mixed citation behavior. Most common role is background (57%).
41
Pith papers citing it
Background
57% of classified citations
abstract
Automating GUI tasks remains challenging due to reliance on textual representations, platform-specific action spaces, and limited reasoning capabilities. We introduce Aguvis, a unified vision-based framework for autonomous GUI agents that directly operates on screen images, standardizes cross-platform interactions and incorporates structured reasoning via inner monologue. To enable this, we construct Aguvis Data Collection, a large-scale dataset with multimodal grounding and reasoning annotations, and develop a two-stage training pipeline that separates GUI grounding from planning and reasoning. Experiments show that Aguvis achieves state-of-the-art performance across offline and real-world online benchmarks, marking the first fully autonomous vision-based GUI agent that operates without closed-source models. We open-source all datasets, models, and training recipes at https://aguvis-project.github.io to advance future research.
hub tools
citation-role summary
background 4
baseline 3
citation-polarity summary
representative citing papers
CUA-Gym generates 32,112 verified RLVR tuples across 110 mock environments, enabling trained models to reach 62.1% and 72.6% on OSWorld-Verified while transferring to WebArena.
ScaleWoB generates 100+ synthetic interactive GUI environments and 1000+ verifiable tasks as web pages, releasing a 120-task mobile benchmark where state-of-the-art agents achieve 27.92% success (17.82% on long-horizon tasks) versus 92.08% for humans, with synthetic results generalizing to real apps
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.
WinDeskGround is a parametrically generated benchmark of 1,356 instruction-target pairs that reveals accuracy declines in state-of-the-art MLLMs under partial occlusion in multi-window GUI settings.
Presents CUActSpot benchmark and renderer-LLM data synthesis that lets a 4B model outperform larger open-source models on complex computer interactions.
DynamicUI improves GUI agent performance in high-dynamic environments by processing interaction videos with frame clustering, action-conditioned refinement, and reflection, outperforming prior approaches on the new DynamicGUIBench spanning ten applications.
SecureWebArena is a new benchmark suite for holistic security evaluation of LVLM-based web agents using diverse simulated environments, attack taxonomies, and multi-layered failure analysis across reasoning, behavior, and outcomes.
Failure-driven self-improvement raises OpenCUA-72B success rate on OSWorld from 42.3% to 48.9% via LLM diagnosis and inference-time code patches, without retraining.
InnerZoom bridges cross-layer evidence in one forward pass to achieve SOTA GUI grounding accuracy on six benchmarks while cutting latency up to 31.8% versus two-pass baselines.
GUICrafter uses curriculum learning on unannotated GUI screenshots for visual grounding followed by RL calibration on limited labels to match or exceed prior GUI agents with far less annotation.
AOHP is an Android-based OS harness that reports 21% higher agent task completion, 52% lower token cost, and improved security compliance through agent-first system primitives.
OpenWebRL trains a 4B visual web agent with online RL on live sites using 0.4K init trajectories and 2.2K RL tasks to reach 67% success on Online-Mind2Web and 64% on DeepShop, outperforming prior open agents.
SCALE introduces three adversarial roles (Selector, Predictor, Judger) and a graph exploration method (SCALE-Hop) to enable MLLM-based web agents to self-discover limitations and improve, backed by the SCALE-20k dataset from 19 websites.
AQuaUI uses adaptive quadtrees to cut visual tokens in GUI-agent LMMs by up to 29.52% at inference time while retaining 99.06% of full-token accuracy on grounding and navigation benchmarks.
DiagEval applies trajectory-conditioned diagnostic probes to recover 45.6-62.1% of misattributed failures in GUI-agent software evaluation, raising accuracy from 69.9% to 78.3% on WebDevJudge-Unit and 65.0% to 81.6% on RealDevBench.
BAMI mitigates precision and ambiguity biases in GUI grounding via coarse-to-fine focus and candidate selection, raising accuracy on ScreenSpot-Pro without training.
SOLAR-RL assigns dense step-level rewards from static trajectory data by detecting first failure points and applying target-aligned shaping to improve long-horizon GUI task completion without full online interactions.
VLAA-GUI adds mandatory visual verifiers, multi-tier loop breakers, and on-demand search to GUI agents, reaching 77.5% on OSWorld and 61.0% on WindowsAgentArena with some models exceeding human performance.
UI-Zoomer uses uncertainty quantification to trigger and size adaptive zoom-ins only on uncertain GUI grounding predictions, yielding up to 13.4% gains on benchmarks with no training.
EchoTrail-GUI builds an automated memory of successful GUI task trajectories via self-exploration and injects relevant past examples to raise success rates on Android benchmarks.
MobiBench reaches near-human offline evaluation fidelity for mobile GUI agents by accepting any valid action at each step, and enables modular attribution of performance to agent components.
MGA is a memory-driven GUI agent that uses an observer for bias-free screen reading and structured memory for compact state transitions to enable efficient long-horizon automation.
RISK introduces a dataset, benchmark, and R1-style RL fine-tuning for GUI agents that achieve 6.8-8.8% offline gains and 70.5% online task success in e-commerce risk management using 7.2% of baseline parameters.
citing papers explorer
-
OSWorld 2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks
On 108 long-horizon real-world computer workflows, frontier agents complete at most 20.6% of tasks and fail mainly by losing hidden state, not by basic GUI control.
-
CUA-Gym: Scaling Verifiable Training Environments and Tasks for Computer-Use Agents
CUA-Gym generates 32,112 verified RLVR tuples across 110 mock environments, enabling trained models to reach 62.1% and 72.6% on OSWorld-Verified while transferring to WebArena.
-
ScaleWoB: Guiding GUI Agents with Coding Agents via Large-Scale Environmental Synthesis
ScaleWoB generates 100+ synthetic interactive GUI environments and 1000+ verifiable tasks as web pages, releasing a 120-task mobile benchmark where state-of-the-art agents achieve 27.92% success (17.82% on long-horizon tasks) versus 92.08% for humans, with synthetic results generalizing to real apps
-
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.
-
WinDeskGround: A Benchmark for Robust GUI Grounding in Complex Multi-Window Desktop Environments
WinDeskGround is a parametrically generated benchmark of 1,356 instruction-target pairs that reveals accuracy declines in state-of-the-art MLLMs under partial occlusion in multi-window GUI settings.
-
Covering Human Action Space for Computer Use: Data Synthesis and Benchmark
Presents CUActSpot benchmark and renderer-LLM data synthesis that lets a 4B model outperform larger open-source models on complex computer interactions.
-
Benchmarking and Improving GUI Agents in High-Dynamic Environments
DynamicUI improves GUI agent performance in high-dynamic environments by processing interaction videos with frame clustering, action-conditioned refinement, and reflection, outperforming prior approaches on the new DynamicGUIBench spanning ten applications.
-
SecureWebArena: A Holistic Security Evaluation Benchmark for LVLM-based Web Agents
SecureWebArena is a new benchmark suite for holistic security evaluation of LVLM-based web agents using diverse simulated environments, attack taxonomies, and multi-layered failure analysis across reasoning, behavior, and outcomes.
-
Learning from Failure: Inference-Time Self-Improvement for Computer-Use Agents
Failure-driven self-improvement raises OpenCUA-72B success rate on OSWorld from 42.3% to 48.9% via LLM diagnosis and inference-time code patches, without retraining.
-
One Forward Beats Two: InnerZoom for Accurate and Efficient GUI Grounding
InnerZoom bridges cross-layer evidence in one forward pass to achieve SOTA GUI grounding accuracy on six benchmarks while cutting latency up to 31.8% versus two-pass baselines.
-
GUICrafter: Weakly-Supervised GUI Agent Leveraging Massive Unannotated Screenshots
GUICrafter uses curriculum learning on unannotated GUI screenshots for visual grounding followed by RL calibration on limited labels to match or exceed prior GUI agents with far less annotation.
-
AOHP: An Open-Source OS-Level Agent Harness for Personalized, Efficient and Secure Interaction
AOHP is an Android-based OS harness that reports 21% higher agent task completion, 52% lower token cost, and improved security compliance through agent-first system primitives.
-
OpenWebRL: Demystifying Online Multi-turn Reinforcement Learning for Visual Web Agents
OpenWebRL trains a 4B visual web agent with online RL on live sites using 0.4K init trajectories and 2.2K RL tasks to reach 67% success on Online-Mind2Web and 64% on DeepShop, outperforming prior open agents.
-
Learning to Adapt: Self-Improving Web Agent via Cognitive-Aware Exploration
SCALE introduces three adversarial roles (Selector, Predictor, Judger) and a graph exploration method (SCALE-Hop) to enable MLLM-based web agents to self-discover limitations and improve, backed by the SCALE-20k dataset from 19 websites.
-
AQuaUI: Visual Token Reduction for GUI Agents with Adaptive Quadtrees
AQuaUI uses adaptive quadtrees to cut visual tokens in GUI-agent LMMs by up to 29.52% at inference time while retaining 99.06% of full-token accuracy on grounding and navigation benchmarks.
-
DiagEval: Trajectory-Conditioned Diagnosis for Reliable Software Evaluation with GUI Agents
DiagEval applies trajectory-conditioned diagnostic probes to recover 45.6-62.1% of misattributed failures in GUI-agent software evaluation, raising accuracy from 69.9% to 78.3% on WebDevJudge-Unit and 65.0% to 81.6% on RealDevBench.
-
BAMI: Training-Free Bias Mitigation in GUI Grounding
BAMI mitigates precision and ambiguity biases in GUI grounding via coarse-to-fine focus and candidate selection, raising accuracy on ScreenSpot-Pro without training.
-
SOLAR-RL: Semi-Online Long-horizon Assignment Reinforcement Learning
SOLAR-RL assigns dense step-level rewards from static trajectory data by detecting first failure points and applying target-aligned shaping to improve long-horizon GUI task completion without full online interactions.
-
VLAA-GUI: Knowing When to Stop, Recover, and Search, A Modular Framework for GUI Automation
VLAA-GUI adds mandatory visual verifiers, multi-tier loop breakers, and on-demand search to GUI agents, reaching 77.5% on OSWorld and 61.0% on WindowsAgentArena with some models exceeding human performance.
-
UI-Zoomer: Uncertainty-Driven Adaptive Zoom-In for GUI Grounding
UI-Zoomer uses uncertainty quantification to trigger and size adaptive zoom-ins only on uncertain GUI grounding predictions, yielding up to 13.4% gains on benchmarks with no training.
-
EchoTrail-GUI: Building Actionable Memory for GUI Agents via Critic-Guided Self-Exploration
EchoTrail-GUI builds an automated memory of successful GUI task trajectories via self-exploration and injects relevant past examples to raise success rates on Android benchmarks.
-
MobiBench: Multi-Branch, Modular Benchmark for Mobile GUI Agents
MobiBench reaches near-human offline evaluation fidelity for mobile GUI agents by accepting any valid action at each step, and enables modular attribution of performance to agent components.
-
MGA: Memory-Driven GUI Agent for Observation-Centric Interaction
MGA is a memory-driven GUI agent that uses an observer for bias-free screen reading and structured memory for compact state transitions to enable efficient long-horizon automation.
-
RISK: A Framework for GUI Agents in E-commerce Risk Management
RISK introduces a dataset, benchmark, and R1-style RL fine-tuning for GUI agents that achieve 6.8-8.8% offline gains and 70.5% online task success in e-commerce risk management using 7.2% of baseline parameters.
-
InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
InternVL3.5 advances open-source multimodal models with Cascade RL for +16% reasoning gains and ViR for 4x inference speedup, with the 241B model reaching SOTA among open-source MLLMs on multimodal, reasoning, and agentic tasks.
-
GTA1: GUI Test-time Scaling Agent
GTA1 combines test-time scaling for action plan selection with RL-based grounding to achieve SOTA results on GUI agent benchmarks.
-
Mobile-R1: Towards Interactive Capability for VLM-Based Mobile Agent via Systematic Training
Mobile-R1 introduces a hierarchical three-stage curriculum that combines format alignment, verifiable action feedback, and multi-turn environment training to improve exploration and self-correction in VLM-based mobile agents, plus a new Chinese GUI dataset and benchmark.
-
UI-R1: Enhancing Efficient Action Prediction of GUI Agents by Reinforcement Learning
UI-R1 shows rule-based RL with GRPO on 136 GUI tasks improves a 3B MLLM's action prediction accuracy by 6-22% over its base model and matches larger SFT-trained models.
-
Fara-1.5: Scalable Learning Environments for Computer Use Agents
Fara1.5 open-weight agents, trained on a new synthetic-plus-live data pipeline, set size-class state-of-the-art results on WebVoyager and Online-Mind2Web.
-
GUI-C$^2$: Coarse-to-Fine GUI Grounding via Difficulty-Aware Reinforcement Learning
GUI-C² pairs a difficulty-scoring data pipeline with an area-gated coarse-to-fine RL mechanism to improve GUI grounding accuracy and training stability.
-
Mobile-Aptus: Confidence-Driven Proactive and Robust Interaction in MLLM-based Mobile-Using Agents
Mobile-Aptus uses supervised fine-tuning followed by semantic similarity retrieval and direct preference optimization to calibrate confidence scores in mobile agents, yielding over 17% average task success improvement on four benchmarks.
-
CaptchaMind: Training CAPTCHA Solvers via Reinforcement Learning with Explicit Reasoning Supervision
Presents CaptchaBench benchmark and CaptchaMind RL solver achieving 82.9% success on benchmark tasks and 71% on real-world CAPTCHAs via explicit reasoning process supervision.
-
SE-GA: Memory-Augmented Self-Evolution for GUI Agents
SE-GA combines Test-Time Memory Extension for dynamic context retrieval with Memory-Augmented Self-Evolution training to reach 89.0% on ScreenSpot and 75.8% on AndroidControl-High.
-
DRS-GUI: Dynamic Region Search for Training-Free GUI Grounding
DRS-GUI introduces a dynamic region search method with Focus/Shift/Scatter actions and MCTS-based planning that improves GUI grounding accuracy by 14% on ScreenSpot-Pro for both general and GUI-specific MLLMs without any training.
-
Measure Twice, Click Once: Co-evolving Proposer and Visual Critic via Reinforcement Learning for GUI Grounding
A co-evolving proposer-critic RL framework improves GUI grounding accuracy by letting the model critique its own proposals rendered on screenshots.
-
Towards Scalable Lightweight GUI Agents via Multi-role Orchestration
LAMO uses role-oriented data synthesis and two-stage training (perplexity-weighted supervised fine-tuning plus reinforcement learning) to create scalable lightweight GUI agents that support both single-model and multi-agent orchestration.
-
What's Missing in Screen-to-Action? Towards a UI-in-the-Loop Paradigm for Multimodal GUI Reasoning
UILoop makes GUI agents reason by cycling through screen, UI elements, and actions, backed by a new 26K-sample UI comprehension benchmark.
-
UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning
UI-TARS-2 reaches 88.2 on Online-Mind2Web, 47.5 on OSWorld, 50.6 on WindowsAgentArena, and 73.3 on AndroidWorld while attaining 59.8 mean normalized score on a 15-game suite through multi-turn RL and scalable data generation.
-
Qwen2.5-VL Technical Report
Qwen2.5-VL reports a vision-language model family using native dynamic-resolution ViT and absolute time encoding that matches GPT-4o on document and diagram tasks while supporting hour-long videos with second-level localization.
-
InfantAgent-Next: A Multimodal Generalist Agent for Automated Computer Interaction
InfantAgent-Next integrates tool-based and vision agents in a modular architecture and reports 7.27% accuracy on OSWorld, exceeding Claude-Computer-Use while also testing on GAIA and SWE-Bench.
-
Large Language Model-Brained GUI Agents: A Survey
A survey consolidating frameworks, data practices, large action models, benchmarks, applications, and research gaps in LLM-brained GUI agents.