REVIEW 7 cited by
AgentCPM-GUI: Building Mobile-Use Agents with Reinforcement Fine-Tuning
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
AgentCPM-GUI: Building Mobile-Use Agents with Reinforcement Fine-Tuning
abstract
The recent progress of large language model agents has opened new possibilities for automating tasks through graphical user interfaces (GUIs), especially in mobile environments where intelligent interaction can greatly enhance usability. However, practical deployment of such agents remains constrained by several key challenges. Existing training data is often noisy and lack semantic diversity, which hinders the learning of precise grounding and planning. Models trained purely by imitation tend to overfit to seen interface patterns and fail to generalize in unfamiliar scenarios. Moreover, most prior work focuses on English interfaces while overlooks the growing diversity of non-English applications such as those in the Chinese mobile ecosystem. In this work, we present AgentCPM-GUI, an 8B-parameter GUI agent built for robust and efficient on-device GUI interaction. Our training pipeline includes grounding-aware pre-training to enhance perception, supervised fine-tuning on high-quality Chinese and English trajectories to imitate human-like actions, and reinforcement fine-tuning with GRPO to improve reasoning capability. We also introduce a compact action space that reduces output length and supports low-latency execution on mobile devices. AgentCPM-GUI achieves state-of-the-art performance on five public benchmarks and a new Chinese GUI benchmark called CAGUI, reaching $96.9\%$ Type-Match and $91.3\%$ Exact-Match. To facilitate reproducibility and further research, we publicly release all code, model checkpoint, and evaluation data.
Forward citations
Cited by 7 Pith papers
-
ProactiveMobile: A Comprehensive Benchmark for Boosting Proactive Intelligence on Mobile Devices
ProactiveMobile is a new benchmark for proactive mobile agents that tests latent intent inference from context and executable API generation, where a fine-tuned 7B model reaches 19.15% success versus 15.71% for o1 and...
-
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.
-
VeriOS: Query-Driven Proactive Human-Agent-GUI Interaction for Trustworthy OS Agents
VeriOS-Agent is an OS agent that proactively queries humans in untrustworthy scenarios via a query-driven framework and three-stage training, achieving 19.72% higher step-wise success rate over baselines while preserv...
-
SparkUI-Parser: Enhancing GUI Perception with Robust Grounding and Parsing
An end-to-end GUI-perception MLLM that uses a token router and continuous coordinate decoder to simultaneously ground and parse interface elements, achieving SOTA on several grounding benchmarks and a new parsing benchmark.
-
Qwen-UI-Agent Technical Report: Toward Next-Generation Real-World Centric Foundation GUI Agents
A real-device-centric foundation GUI agent with hybrid GUI+CLI batched actions, AutoResearch data flywheel, online RL, and a proactive harness reaches SOTA mobile and competitive desktop/web scores.
-
Where Not to Learn: Prior-Aligned Training with Subset-based Attribution Constraints for Reliable Decision-Making
Penalizing off-prior attribution evidence during training with subset-selection attribution improves accuracy and attribution reasonability in image classifiers and MLLM-based GUI agents.
-
Where Not to Learn: Prior-Aligned Training with Subset-based Attribution Constraints for Reliable Decision-Making
A training approach that adds attribution constraints derived from human priors to steer models toward using specified input regions for decisions.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.