jacob-valdez/synthux-economy-r5-w054
收藏资源简介:
SynthUX计算机使用数据集 — 经济模拟r5是一个基于模拟环境的计算机使用轨迹数据集,由SynthUX生成。每个记录代表一个模拟公司内工人的设备会话:目标扩展为节点树,终端节点通过低级鼠标/键盘输入驱动真实桌面模拟器应用(如终端、笔记、VS Code、浏览器、Slack/Teams、邮件、表格、幻灯片等),观察到的轨迹以屏幕视频和每个节点的帧形式记录。数据集支持应用原生的多应用程序执行(无工作台覆盖),每个参与者的内容多样性,跨操作系统消息传递,以及大型多公司经济模拟。每个经济模拟作为独立的数据集仓库发布。数据集包含data.jsonl文件(每个渲染轨迹一个记录)、screenshots目录(每个轨迹的每个终端节点在最后输入事件时捕获的帧)和videos目录(完整的屏幕录制)。记录模式包括轨迹ID、参与者ID、公司、角色ID、标题、环境(模拟器类型)、自然语言目标、节点树、密集的低级输入事件、观察结果、对齐信息以及媒体文件。本构建包含36个轨迹、1090帧和6个公司,按环境分布为browser-os(12)、macos-web-next(15)和windows-web-next(9)。数据集使用SynthUX生成,观察结果是因果的(突变通道为低级输入),捕获过程中未使用高级应用程序突变。
The SynthUX Computer Usage Dataset — Economic Simulation r5 is a simulation environment-based computer usage trajectory dataset generated by SynthUX. Each record represents a device session of a worker within a simulated company: Goals are expanded into node trees, and terminal nodes drive real desktop emulator applications such as terminals, note-taking apps, VS Code, browsers, Slack/Teams, email clients, spreadsheets, slides, etc. via low-level mouse and keyboard inputs. Observed trajectories are recorded as screen videos and frames for each node. The dataset supports natively multi-application execution (no workspace overlay), content diversity per participant, cross-operating system messaging, and large-scale multi-company economic simulations. Each economic simulation is released as an independent dataset repository. The dataset contains a data.jsonl file (one record per rendered trajectory), a screenshots directory (frames captured at the final input event for each terminal node of every trajectory), and a videos directory (full screen recordings). The record schema includes trajectory ID, participant ID, company, role ID, title, environment (emulator type), natural language goal, node tree, dense low-level input events, observations, alignment information, and media files. This build contains 36 trajectories, 1090 frames, and 6 companies, distributed across environments as browser-os (12), macos-web-next (15), and windows-web-next (9). The dataset is generated using SynthUX, where observations are causal (the mutation channel is low-level inputs), and no high-level application mutations were used during the capture process.




