Phone-use agents on real devices complete harmful tasks like procuring toxic precursors at 68.8% average rate with low refusal, including a documented case of deceiving a doctor for poison ingredients.
hub Mixed citations
Mobile-agent-v3
Mixed citation behavior. Most common role is background (57%).
29
Pith papers citing it
Background
57% of classified citations
hub tools
citation-role summary
background 4
baseline 2
method 1
citation-polarity summary
years
2026 29representative citing papers
MobileForge adapts Qwen3-VL-8B to 67.2% Pass@3 on AndroidWorld using only automatically generated annotation-free data via MobileGym and HiFPO, with ForgeOwl-8B reaching 77.6%.
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.
Introduces LivingScreen benchmark for living-screen-native GUI agents on short-video platforms; frontier models fail to match human cost-accuracy due to over- and under-observation.
AndroidDaily supplies 350 verifiable tasks on 94 closed-source Android apps evaluated by GRADE (87.37% human agreement), with the strongest model achieving 62% success.
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
Presents CUActSpot benchmark and renderer-LLM data synthesis that lets a 4B model outperform larger open-source models on complex computer interactions.
GUI grounding in VLMs is bottlenecked by prefill-stage candidate selection that decoding cannot fix, so Re-Prefill uses attention to extract and re-inject target tokens for up to 4.3% gains on ScreenSpot-Pro.
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.
RiskWebWorld is the first realistic interactive benchmark for GUI agents in e-commerce risk management, revealing a large gap between generalist and specialized models plus RL gains.
Introduces Active Task Driving Memory (ATMem) and STR-GRPO to move GUI agents from passive record storage to actively maintained task states, tested on a new mobile benchmark with progress and scope-aware metrics.
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.
PhoneBuddy combines real-app and mock-app RL after shared SFT, raising real-phone task success from 36.67% to 45.33% and AndroidWorld from 60.3% to 83.2%.
MemGUI-Agent uses Context-as-Action (ConAct) for proactive context management in long-horizon GUI tasks, trained on the MemGUI-3K dataset to achieve top 8B-model results on MemGUI-Bench and MobileWorld.
SGCD generates supervision for off-trajectory states in GUI agents by mixing expert trajectories with continuations produced by a skill-guided policy after the base policy reaches those states.
CLI-based coding agents outperform GUI baselines on AndroidWorld and MobileWorld, with oracles reaching 88.8% and 86.3% solvability and a new CLI-Advantage suite showing CLI superiority in bulk operations, filtering, aggregation, cross-app workflows, and hidden state.
DeskCraft provides 538 tasks across design, video, audio, and 3D software with a multilevel taxonomy and formalized mid-turn and post-turn human-agent interaction protocols, evaluating 18 agents with top performance at 31.6% on standard tasks.
PhoneWorld is a pipeline that converts real mobile trajectories into scalable controllable environments, yielding large gains on four benchmarks when used to supplement training data.
LearnWeak specializes small CUAs via weakness detection by a reference agent, targeted task synthesis, and error-aware training, delivering 11+ point gains on OSWorld.
OpenComputer introduces a verifier-grounded framework with state verifiers, self-evolving layers, task synthesis, and auditable evaluation for 33 desktop apps and 1000 tasks to support computer-use AI agents.
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.
MementoGUI introduces a modular memory-control framework with working and episodic memory operators that improves long-horizon GUI agent performance over history-replay and text-only baselines.
ToolCUA introduces a trajectory scaling pipeline and staged RL to optimize GUI-tool switching, reaching 46.85% accuracy on OSWorld-MCP for a 66% relative gain over baseline.
Phone-use agents avoid harm more often through inability to act than through deliberate safe choices, so benchmarks must separate unsafe judgment from capability failure.
citing papers explorer
-
What Happens Before Decoding? Prefill Determines GUI Grounding in VLMs
GUI grounding in VLMs is bottlenecked by prefill-stage candidate selection that decoding cannot fix, so Re-Prefill uses attention to extract and re-inject target tokens for up to 4.3% gains on ScreenSpot-Pro.
-
Beyond the GUI Paradigm: Do Mobile Agents Need the Phone Screen?
CLI-based coding agents outperform GUI baselines on AndroidWorld and MobileWorld, with oracles reaching 88.8% and 86.3% solvability and a new CLI-Advantage suite showing CLI superiority in bulk operations, filtering, aggregation, cross-app workflows, and hidden state.
-
MementoGUI: Learning Agentic Multimodal Memory Control for Long-Horizon GUI Agents
MementoGUI introduces a modular memory-control framework with working and episodic memory operators that improves long-horizon GUI agent performance over history-replay and text-only baselines.