已审阅并标注「可观测性」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。
MAP annotates computer-use traces at linked operation, function/consequence, method and goal levels, then compares human and agent workflows through sequence, pattern and effort metrics.
MobileInject evaluates visual/textual observation injection in stateful Android workflows using resettable emulators, benign/adversarial state oracles and trace-based diagnosis of emitted rationales and actions.
MobileJailBench evaluates mobile-agent jailbreaks in a browser-based Android-like simulator, combining feedback-driven attack evolution, separately admitted task expansions and evidence-grounded outcome adjudication.
Measures source-attribution limits in Android accessibility and separately tests application-history and selected-WebView-object checks for authorizing AppFunctions actions.
Evaluates duration following, runtime forecasts and elapsed-time estimates in native agent harnesses, including OSWorld desktop tasks, with external timers, native graders and work-versus-wait analysis.
Describes WebFovea, a vision-based live-web agent whose execution loop checks action parsing, browser effects, feedback accuracy, and observation delivery. The report analyzes failures such as coordinate mismatches and silent control interactions and documents corresponding harness changes.
Audits WebArena-Lite outcomes and trajectories with human review to identify automatic-evaluator false negatives and the first consequential execution errors. It also studies explicit execution-state memory and procedural guidance, showing why final task scores can miss substantial differences in agent behavior.
Improves GUI-agent execution code by analyzing action effects and repeated-task failures, then validating bounded changes with frozen model weights.
Studies repetitive, unproductive actions in desktop agents through their internal activation states. It proposes Reset, Reroute, Restore, a context intervention that temporarily resets an agent’s trajectory and later restores it to help escape repeated ineffective behavior.
Introduces GUITAR, which groups visually different screens into functional states and analyzes agent failures over a state-transition graph. It uses the resulting bottlenecks to diagnose recurring mobile and web errors and evaluate targeted guidance.
Adds annotated intermediate subgoals to computer-use tasks and uses process-oriented evaluation to identify action and progress failures.
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.
Introduces CURA, a telemetry-based runtime monitor that turns computer-use trajectories into sequential failure alarms and can trigger mid-task oversight. Its certificate controls false alarms, while the study separately reports detection performance and limits of retrospective monitoring.
Audits failure-scored computer-use trajectories to distinguish agent errors from broken tasks and evaluator false negatives. It organizes reliability issues across task construction, trajectory evidence, scoring, and reporting, then derives corresponding evaluation design practices.
Introduces Sidekick, a communication interface that uses ambient status cues, multimodal action summaries, and visual and verbal explanations of reasoning to help people monitor computer-use agents. A user study examines progress awareness, error traceability, and multitasking across background and foreground interaction.
Plover exposes GUI-agent plans and replanning as persistent artifacts that users can inspect and edit during execution. Its planner–executor design supports localized correction through plan edits, language guidance and screenshot-grounded interventions, studied through failure repair and workflow scenarios.
Tactile turns accessibility, OCR and visual evidence into explicit desktop targets with executable actions and verification cues. Its execution loop chooses semantic actions or coordinate fallbacks while retaining provenance for replay and diagnosis.
WebRetriever evaluates navigation, knowledge-assisted interaction and end-to-end information-extraction workflows under separate protocols. Its NavEval judge uses richer interaction context to diagnose cases where navigation success does not translate into useful task completion.
Evaluates web agents against anti-bot mechanisms on instrumented honeysites and analyzes network, HTTP and browser fingerprints. The study compares resistance to blocking with the ability to distinguish agents, humans and agent implementations.
Builds a logging framework that combines network/protocol fingerprints with browser interaction behavior to attribute web-agent traffic. A decision-tree classifier is evaluated across six agent implementations, human browsing and conventional crawlers.
HANSEL extracts evidence pages and snippets from browser-agent traces and preserves relevant page state in interactive verification views. It flags untraceable answers and evaluates evidence selection and human verification effort separately.
Syll combines tool calls, shell execution and visual GUI control in a personal-automation runtime. It turns user demonstrations into reusable skills and exposes execution logs, keyframes and approval checkpoints for inspection.
Introduces OpenComputer, which builds desktop tasks around application-state verifiers and refines those verifiers using execution feedback. An evaluation harness records trajectories and assigns auditable partial credit across professional software workflows.
AgentClick gives terminal-based agents a browser review interface for plans, drafts, execution traces and editable memory. A local server and skill plugin let users inspect proposed actions and provide structured feedback.
This position paper derives interface design principles from terminal-based human-agent collaboration: representational compatibility, transparent actions and low participation barriers. It argues that graphical and other agent-facing modalities should provide these properties explicitly.
CUAAudit examines vision-language models that judge computer-use task completion from an instruction and final environment state. Across macOS, Windows and Linux benchmarks, it measures accuracy, confidence calibration and disagreement between auditors.
MobiBench evaluates mobile GUI agents offline while allowing multiple valid action paths. It decomposes agent pipelines into modules to study component contributions, configurations, and performance bottlenecks.
AUI-Gym evaluates generated websites by having computer-use agents attempt tasks whose executability is checked programmatically. A feedback dashboard condenses navigation histories into visual summaries that a coding agent can use to revise an interface.
ProBench extends mobile GUI evaluation with tasks whose correctness depends on the interaction process as well as the final state. Its ProcessProvider supplies process information for assessing agent behavior and analyzing failure modes.
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.
Presents an agent-engineering framework combining runtime goal clarification, an observable component architecture and evolving execution memory. The Fairy mobile GUI agent evaluates these mechanisms on ambiguous tasks and through controlled and user studies.
Introduces a modular framework for identifying errors at intermediate stages of a web-agent pipeline. A SeeAct and Mind2Web case study examines weaknesses that are obscured by final task-success metrics.
Introduces AgentSentinel, which intercepts sensitive computer-agent operations and audits them using task context together with system traces. BadComputerUse evaluates this enforcement approach across attack scenarios involving tool and computer control.
RealDevWorld evaluates generated software repositories with 194 development tasks and AppEvalPilot, an agent judge that operates the resulting GUIs. The evaluation examines functional behavior, visual fidelity, and runtime responses to produce task-specific diagnostic feedback.
FineState-Bench evaluates fine-grained GUI perception and control across desktop, web, and mobile interfaces. Its Visual Diagnostic Assistant separates perception and localization contributions in controlled interaction experiments.
OSWorld-Human adds human-determined trajectories to OSWorld to evaluate the efficiency of desktop computer-use agents. The study compares 16 agents and analyzes model-call latency, changes in step duration during task execution, and excess action steps.
WebSuite organizes browser operations into an action taxonomy and tests both isolated controls and complete workflows. Its diagnostic tasks attribute failures to specific actions, enabling comparison of text-based and multimodal web agents.
Introduces AgentBoard, a multiturn agent evaluation framework with intermediate progress metrics and analysis tools. Its broader task suite includes WebShop and WebArena interactions with stepwise progress measurements.
Local, cloud-free replay debugger that records each Browser Use step's screenshot, URL, action and model I/O as a timeline and reports the step at which a failed run diverges from a known-good one.
Optimizes GUI-agent execution code from repeated rollout evidence and checks proposed changes against fixed validation gates.
A screenshot-driven browser agent over Chrome DevTools Protocol, with resumable sessions, context compression, action replay and configurable completion checks.
Inspect-based multi-agent evaluation framework with role, scheduling and communication controls, BrowserART accessibility-tree actions, and desktop mouse, keyboard and screenshot tools.
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。
