已审阅并标注「强化学习」的研究与资产。
条目依据已审阅标签归类，同一主题不表示资源之间存在依赖关系。日期保留来源精度，注释保留原语言。
WildAWM synthesizes resettable GUI/MCP/CLI environments and cross-interface tasks, using execution-based checks of solvability, difficulty and verifier consistency before supervised and reinforcement learning.
VeriCUA trains a Judge to select verification checks and stopping decisions while a frozen Probe inspects the post-task GUI. Single-attempt reinforcement learning balances verdict correctness and Probe use.
CUA-Lite unifies environment interfaces, supervised data formats and per-model harnesses for evaluation, reinforcement learning and supervised fine-tuning across desktop, browser and mobile tasks.
MAGE retrieves software-manual sections into fixed execution plans and generates setup/main task-plan pairs whose GUI rollouts support application-specific policy adaptation with round-based GRPO.
Trains native computer-use models to coordinate GUI actions and code execution using hybrid demonstrations and online reinforcement learning.
ECHOES learns recursive textual task memory through privileged screenshot-history distillation and GRPO, then predicts GUI actions from the current screenshot and memory.
Unifies GUI evaluation, trajectory collection, environment synthesis, and online reinforcement learning across mobile, computer, and web tasks.
Extends computer-use policies to game control through synthetic training environments and richer keyboard and pointer actions. Its evaluation concerns games rather than ordinary productivity workflows.
Introduces ScreenRef screenshot grounding data and a training sequence from supervised learning through teacher-guided distillation to reinforcement learning for GroundGUI-8B.
Synthesizes resettable workspaces with files, application observations and skills; repairs environment incompatibilities and trains long-task agents using verifier-linked turn credit.
Groups GUI trajectories by subgoal and combines action-level and group-level assessments into intermediate rewards for training agents with sparse task outcomes.
Introduces LatentGUIWorld to test learning interface semantics and dynamics through interaction, and studies finer movement actions with intermediate visual feedback and exploration demonstrations.
Introduces OpenMobile-2, combining simulated mobile apps, emulator tasks, released training data, and policies that can alternate GUI interaction with app-native tools.
Presents AdvSim2Real, which jointly adapts web-task curricula, prompt-injection adversaries, and agent policies inside a frozen web world model. It evaluates robustness against an unseen adversary and transfer from simulated interaction to a real browser.
Introduces PAIR for controlled evaluation of mobile GUI agents across user-conditioned application states and RePAIR for training from cross-user subgoal outcomes. The study examines failures when personalized interfaces change which items an agent must identify.
Introduces AMBER, an append-only memory framework in which a web agent learns to reason, act, and retain evidence using outcome-reward reinforcement learning. It studies whether preserving entries avoids the loss of facts and corrective feedback seen with rewritten memory.
Introduces CLIFT, which certifies agent-generated verification signals against training-time judges and reuses them for per-step rewards and trajectory selection. At inference, a frozen question bank helps compare candidate web rollouts without querying an external judge.
Introduces ComputerSD, which turns feedback from executed GUI transitions into online self-distillation guidance and step-level scores. A GUI analyzer supplies these signals, which are combined with trajectory-level reinforcement learning in asynchronous training.
Trains computer agents on verified GUI, command-line, and interleaved trajectories to learn when and how to combine interaction modes.
Introduces Spatial Credit Assignment, which uses click coordinates to refine group-relative reinforcement-learning updates for GUI grounding. It distinguishes failed clicks by spatial evidence while leaving the deployed action policy unchanged.
Introduces CUA-Sandbox, which separates each rollout’s mutable state from shared initialized software runtimes. State-scoped execution and transactional resets or branches aim to reduce training-environment costs while retaining original interfaces and evaluators.
Introduces CaptchaArena, a dataset of execution-verified interactive CAPTCHA solutions with screenshot–action trajectories, reasoning annotations, and pixel masks for irregular targets. The work trains a single policy across CAPTCHA types using supervised learning followed by verifier-reward reinforcement learning.
GUI-Hopper combines verified Android deeplinks with GUI actions, using landing-screen descriptions, hybrid trajectory training and reinforcement learning to select navigation shortcuts and recover through GUI control.
Trains a mobile planner with supervised trajectories and competence-adaptive RL, then couples typed tool execution with memory, skills and offline model-harness refinement.
MintAct trains a visual-agent family for grounding, navigation and tool use across mobile, desktop and web environments. Its shared infrastructure supports trajectory collection and asynchronous online RL while controlling cross-domain sampling and off-policy drift.
Describes a mobile-agent training pipeline centered on real-device rollouts, recovery of useful supervision from failures, and evolving evaluation tasks.
Introduces Iron, a GUI-agent training framework that aligns low-level actions with high-level intent using cycle-consistency rewards. It also repurposes failed trajectories through hindsight reconstruction and evaluates transfer to other digital tasks.
Introduces SCOPE, which trains computer-use agents on ordinary task demonstrations, safe continuations around hazards, and explicit refusals. Its SCOPE-Gen pipeline pairs verifiable tasks with environmental-risk variants to support joint capability and safety training.
Describes a GUI foundation agent trained through expanded environments, function-grounded tasks, and trajectory verification with safety-aware execution mechanisms.
Uses learned world models for mobile-agent training rollouts and for reasoning about the consequences of candidate GUI actions.
ASIL exposes application state as structured JSON and provides code-executable semantic actions. Its application benchmark compares this interface with screenshot control and application-native APIs, and studies supervised and reinforcement learning on the structured interface.
Introduces AdaptRubric, which retrieves reusable criteria for a GUI task family and then generates instruction-specific checks for concrete values and constraints. It evaluates these adaptive rubrics in offline reward assessment and online agent optimization.
Introduces AnTrap, which injects task-solvable runtime anomalies into Android GUI-agent trajectories using a layered failure taxonomy. It compares model robustness and adversarial reinforcement learning to distinguish recoverable interface traps from persistent contextual failures.
Introduces GSAR, which synthesizes GUI tasks and successful goal states, then annotates task-relevant interface elements as reward anchors. These anchors support offline trajectory verification and online policy training.
Builds synthetic websites with verified navigation, backend state, and feasible tasks to provide reliable supervision for web-agent training.
Introduces LACL-GUI, a contrastive learning objective that distinguishes trajectory quality within both successful and failed GUI executions. It favors concise successes and uses divergence from successful trajectories to differentiate failures during policy optimization.
Combines a browser execution framework, recovery-oriented supervision, and online reinforcement learning, with a bilingual live-web task benchmark.
LookAgain treats a GUI coordinate as a revisable hypothesis, rendering its marker and a local screenshot patch before confirmation. Reflective supervised trajectories provide a cold start, followed by GRPO using terminal grounding correctness.
AppDeltaWorld predicts executable code updates for the next mobile UI state, using transition-constrained retrieval and rendered visual assets. It evaluates reconstruction fidelity separately from policy training and adaptation using generated environments.
This study separates tool-adoption behavior from task competence in hybrid GUI–MCP agents. It tests tool-use incentives and screenshot-history compression, showing why changed action preferences alone need not improve held-out accuracy.
Qwen-CUA operates through screenshots and keyboard/mouse events while compressing older visual history into reusable context blocks. Its report combines verifiable task generation, trajectory-based training and separate browser and Bash-augmented studies.
Describes a foundation GUI agent with combined GUI and command-line actions, large-scale interaction infrastructure, and iterative environment and task generation.
Introduces Echoverse, which compiles specifications into stateful applications with database-grounded graders and uses rollouts to improve both the environments and agent policies. It examines behavioral depth, targeted interface practice, and environment repair as sources of training value.
OpenForgeRL connects stateful agent harnesses to reinforcement-learning infrastructure through a model-call recording proxy and isolated rollout containers. Its validation includes browser and computer-use agents, alongside tool-oriented agents, and examines how harness choice affects learned reliability.
Studies a controlled setting where GRPO fails to improve a strong small web agent and can reduce text-track performance. Matched reachable-reward tasks and component grafting distinguish limited optimization headroom from learning-rate-dependent failure mechanisms.
Scales computer-agent reinforcement learning through verifiable task generation, capability-aware rollout allocation, and segmented visual context.
MAG jointly evaluates browser task actions and user-facing guides grounded in screenshots through element marks or pixel coordinates. Its harness covers annotation, live evaluation and joint metrics, with an expert-trajectory-augmented GRPO training study.
EvoCUA-1.5 trains multi-turn desktop agents through online reinforcement learning in executable sandboxes. It combines step-level policy optimization, task filtering and calibration, an adaptive curriculum, and asynchronous rollout handling with staleness control.
Describes a mobile GUI agent trained through real-device interaction, corrected failure trajectories, supervised learning, and reinforcement learning.
GUICrafter first learns grounding from large collections of unannotated screenshots and webpages, then uses a smaller high-quality dataset for reinforcement-learning calibration. The curriculum targets reduced dependence on manually annotated GUI training data.
WinDOM obtains bounding-box supervision from a Windows 11 web reimplementation driven through Playwright, and studies rejection-sampling distillation for small GUI grounders. It treats the depth of the supervised cold start as a GRPO choice, comparing same-size EMA and larger same-family teachers.
The framework uses a vision-language judge’s final-screen assessment as a terminal reward for GUI reinforcement learning. It models evaluator noise and corrects the reward estimator used by PPO across desktop benchmarks.
ENVS branches over GUI actions in live desktop VMs, verifies successful leaves, and converts them into balanced step-level training supervision. It also introduces OSWorld-Noisy to test recovery from dynamic interruptions while retaining the underlying tasks.
Segments and clusters GUI trajectories into inspectable skills, then trains a skill-aware policy from those annotations. Its diagnostic results distinguish readable source-domain clusters from reliable transfer and report limited or absent downstream gains.
MobileForge couples real-app exploration and generated tasks with hierarchical feedback for mobile-agent adaptation. Its optimizer turns trajectory outcomes, step feedback and corrective hints into contextualized GRPO updates, with evaluation on AndroidWorld and a GUI-only MobileWorld split.
VISTA forms GRPO comparison groups from target-preserving crops of the same GUI instance with exactly remapped target boxes. A conditional, advantage-weighted cross-view anchor supplies additional supervision without entering the group baseline, and evaluation includes view-consistency robustness.
GUI-AC uses grounding certainty to adjust advantage weighting and policy-clipping bounds during continual reinforcement fine-tuning. Its mechanisms aim to suppress noisy confidence updates while retaining exploration under shifting GUI domains and resolutions.
AliyunConsoleAgent trains browser agents to verify documented procedures against changing cloud-console UIs through trajectory distillation and online GRPO. Provisioned rollout environments and backend audit-log checks support training and outcome evaluation.
StainFlow tracks visually verifiable task entities through GUI trajectories to identify phases and retrieve focused evidence windows. Its process-reward model is tested for both online policy learning and trajectory-completion judgments.
AsyncWebRL overlaps web-agent rollouts, updates and policy refresh while managing screenshot handling and a persistent rollout pool. It also replaces trajectory-length normalization in multi-step GRPO to reduce verbose failed trajectories, with throughput and out-of-distribution evaluation.
第 1–60 条，共 242 条资源。
通过主题导出清单获取完整记录和证据。克隆目录后，可将主题记录直接交给标准 RAG 与引用导出工具。
