已审阅并标注「交互历史」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。
ScopeMem conditions latent-memory prefixes on a frozen GUI agent's initial action and trains them to correct errors while preserving correct decisions. Its evaluation separates action repairs from harmful revisions.
FocusGUI learns step-specific weights over screenshot, audio, video and interaction-context channels, injecting a compact residual into a frozen decoder without pruning its inputs.
ECHOES learns recursive textual task memory through privileged screenshot-history distillation and GRPO, then predicts GUI actions from the current screenshot and memory.
Surveys multimodal memory using representation and management axes, distinguishing fixed-policy updates from policy evolution and comparing GUI memory evaluation with other modalities.
Pairs factual probes with next-action choices on frozen agent traces, including Mind2Web, to diagnose reading-decision gaps without closed-loop execution.
GUI-SD-v2 extends on-policy self-distillation from grounding to multi-turn GUI interaction. It first strengthens responses to privileged guidance, then selectively distills step-specific reasoning and memory support for later actions.
AnchorGUI compares expected and observed UI transitions and selectively retains screenshots when they disagree. This asymmetric memory supports immediate correction and focuses cross-attempt experience distillation on likely failure steps.
TRACE orders screenshot evidence for monotonic visual-token reduction using layout priors, instruction relevance and feature novelty. Spatial token reserves and incremental KV contraction aim to preserve useful evidence for future GUI targets without repeatedly encoding retired frames.
GraphDroid combines clustered memory of visited GUI states with asynchronous test-intent generation. A hybrid executor reserves LLM calls for complex intents and uses heuristics for simpler ones, evaluated through Android app coverage, bug discovery and crash detection.
Predicts web-agent failure risk from observable trajectory prefixes using cross-step behavior and black-box consistency queries. Key-step labels preserve valid early prefixes of failed runs, and evaluation tests early intervention and transfer across website categories.
Task Model Induction disentangles concurrent activities in screenshot and mouse/keyboard traces, then constructs hierarchical goals and procedural control flow for each task. Evaluation tests reconstruction of observed execution and reuse of derived skills on held-out tasks.
Gated Hindsight Distillation gives a teacher access to the next screenshot during training while the student sees only the available trajectory prefix. Distillation is applied when hindsight repairs a student failure, and the resulting policies are evaluated on Android tasks.
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.
KnowAct-GUIClaw combines a host planner with pluggable GUI subagents, attributable experience memory and an evolving skill library. It uses user profiles and feedback for task decomposition and evaluates transfer across Android, iOS, HarmonyOS and Windows.
MemGUI-Agent makes context management an action chosen by the same policy that controls the mobile UI. Structured history, UI-state and recent-step fields are trained from annotated trajectories to retain cross-app facts while limiting context growth.
This study finds that full-image visual memory can reduce some state errors while worsening action-level failures. Its AGMem alternative stores local crops associated with successful actions or recovery and evaluates their use in desktop GUI tasks.
HiViG combines a compact history of completed actions with a visually grounded critic that checks proposed coordinates before execution. Its evaluations separate history-related planning failures from errors caused by targeting the wrong screen element.
State-Grounded Dynamic Retrieval extracts reusable sub-procedures from completed web trajectories and represents them as paired text and code. Skills are retrieved against both the current webpage state and the task goal at intermediate execution steps.
STaR-KV compresses GUI-agent KV caches by calibrating visual-token importance across attention subspaces and interaction time. Its training-free scoring combines spatial specialization, temporal redundancy and an adaptive temperature, evaluated at matched cache budgets.
Claw-Anything benchmarks persistent personal assistance across activity histories, backend services and integrated GUI/CLI interaction. Its simulated event histories introduce irrelevant or conflicting context, and a separate generation pipeline supplies training environments.
MementoGUI learns a controller for selecting, compressing and retrieving textual and region-level visual memories while leaving the GUI-agent backbone fixed. It also converts computer-use traces into controller training data and evaluates semantic actions, progress and memory consistency on long-horizon tasks.
SE-GA retrieves hierarchical memories during inference and uses the collected experience in a separate policy-improvement pipeline. Its evaluation includes offline grounding/action tasks and online Android interaction, which measure different capabilities.
LongMemEval-V2 asks memory systems to recover environment knowledge from long web-agent histories for question answering. Its AgentRunbook variants contrast indexed knowledge pools with coding-agent retrieval from stored trajectory files.
Trains GUI agents with a learned selector that removes repeated visual patches across consecutive screenshots while preserving spatial structure. ReVision evaluates this history representation on OSWorld, WebTailBench and AgentNetBench.
Proposes a step-level model cascade that runs a smaller computer-use policy until learned monitors flag stalled progress or semantic drift. Separate history-based stuck detection and milestone checks determine when to escalate for recovery or verification without retraining the larger model.
Studies visual token pruning for GUI navigation, separating reusable screenshot encoder embeddings from later language-model computation. The analysis examines foreground-background token allocation and spatial coverage when pruning historical screenshots.
MemoryArena links tasks across sessions so that agents must distill earlier interaction experience and use it in later decisions. Its evaluation includes web navigation alongside planning, search and formal reasoning, testing memory through subsequent actions rather than recall alone.
CCPO couples GUI policy optimization with coordinate-guided compression of past visual observations. It aggregates targets across rollouts to focus retained context and uses distance-based advantages to train grounding and compression together.
Introduces AgentProg, organizing mobile-agent interaction history as a program with variables and control flow to decide what context to retain. A global belief state handles partial observability, with evaluation on AndroidWorld and extended long-horizon tasks.
HiconAgent trains GUI policies to use history through variable-length context sampling and paired policy updates with compressed and complete histories. Its compression retains past actions while removing historical observations, and an alignment loss couples the two branches.
Introduces WebCoach, which condenses navigation logs into persistent episodic memory and retrieves experiences by similarity and recency. Runtime hooks selectively supply advice to browser agents, evaluated on WebVoyager without retraining the underlying models.
HAR-GUI combines episodic reasoning and short-term memory to use earlier interactions when selecting GUI actions. Its training constructs reflective scenarios and correction guidance and applies a hybrid reinforcement-learning reward to the 3B policy.
Combines structured progress-and-decision reasoning with history summarization to improve cross-domain generalization in GUI navigation agents.
Introduces WebGraphEval, which merges multiple agents' trajectories into a weighted graph of canonical actions. Structural analyses examine recurring behavior, inefficient paths and important decision points beyond a final success score.
Encodes GUI trajectories into compact continuous embeddings that can be retrieved directly into an agent’s input. An automated environment-discovery and rollout pipeline expands this memory for long tasks and unfamiliar interfaces.
Introduces JEF-Hinter, which distills successful and failed trajectories into compact episodic hints and retrieves them for the current agent state. Web benchmark experiments examine guidance from decisive steps rather than full demonstration replay.
GUI-KV compresses vision-language agents’ attention caches using visual-token saliency and redundancy between screenshot frames. The method requires no retraining and evaluates how cache budgets affect GUI action accuracy and computational cost.
PAL-UI combines observation/action summaries with selective retrieval of historical screenshots during planning. It trains mobile navigation models from step-level supervision and evaluates transfer to web navigation without additional training.
SimpAgent prunes distracting interface elements with visual masks and compresses interaction history using consistency-based guidance. The study evaluates these two forms of context reduction during GUI navigation in web and mobile environments.
FingerTip 20K gathers 20K human Android demonstrations from users' long-term everyday app use, with user context, to evaluate proactive task suggestion and preference-aware task execution.
UI-Genie jointly improves a mobile GUI agent and a reward model that processes screenshots with interaction history and supplies action- and task-level feedback. Its data pipeline uses rule-based verification, corrupted trajectories, and hard negatives to support reward-guided exploration and iterative training.
Studies backdoor risks in mobile GUI agents through AgentGhost, which combines goal-level and interaction-history triggers during model training. The paper evaluates multiple mobile agents and a tailored defense.
ScaleTrack combines a unified collection of GUI grounding samples with a planning objective that predicts the next action while reconstructing prior actions. This backward prediction links the current screenshot to interaction history during agent training.
Analyzes how interaction history and webpage representation affect multiturn web-agent generalization across unfamiliar websites, task categories, and geographic settings.
LiMAC combines a compact Action Transformer with a fine-tuned vision-language model for Android app control. It uses textual goals and histories of screenshots and UI trees to predict actions, with evaluation on two mobile-control datasets.
Dynamic Planning of Thoughts revises a mobile GUI plan using current environmental feedback and execution history. The study compares this planning strategy with long-dialogue prompting and examines its behavior on unfamiliar tasks and different language-model backbones.
Defines goal identification from observed UI trajectories and an evaluation method for comparing intent descriptions within the same interface context. Experiments reuse Android and web automation data to compare human and model inference of users’ detailed goals.
Trains a smartphone assistant from sequences of screenshots paired with earlier actions. The model predicts UI-only interactions such as taps and swipes, with evaluation on Android in the Wild.
AllTogether standardizes prompts for HTML-based web navigation and compares prompting with instruction fine-tuning. Its analysis examines how HTML context, interaction history, and current environmental feedback affect task execution.
Describes Eager, which detects repeated HyperCard interactions, generalizes them into loops, and previews predicted actions by highlighting interface objects. Users can then ask it to execute the inferred repetitive task.
The context-aware Aria-UI release that uses interaction context to support grounding in dynamic Android and desktop agent tasks.
MemGUI-Agent training and offline evaluation tools provide trajectory conversion, LoRA SFT templates, and step-level checks of GUI actions, memory updates and context folding.
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。
