Agent World Model in the Wild: Computer-Use Agent Environment Synthesis
Frames environment synthesis as a way to avoid the risks of learning directly in real computer systems.
AI Storyline
4 items · 2 sources · 4 days
Operational story trace
Environment Synthesis
Follow in this browser to see new updates on your Live feed.
Current stateDeveloping
Latest change
Agent-R² (Nov 5) synthesizes training data from real sources to target realistic productivity work, the fourth environment-synthesis paper in 17 days.
Earlier contextThe story so far
In October, papers began synthesizing training environments for agents instead of collecting human demonstrations, first for computer use and then for tool use with controllable data composition. By November the work moved to MCP tool-centric synthesis and to real-source data for productivity tasks.
editor-curated · source-linked
Arc
Oct 19Nov 5 · now
START · Oct 19Computer-use agents learn from synthesized environments1 source · show source ▾
SCALING · Oct 27Tool-use scaling laws tied to environment data composition1 source · show source ▾
Scaling Laws for Tool-Use Agents via Controllable Environment Synthesis
Studies how data composition in synthetic tool-use environments affects agent performance.
THE TURN · Nov 3ToolSpace makes MCP environment synthesis tool-centric rather than data-centric1 source · show source ▾
ToolSpace: Scaling Tool-Centric Environment Synthesis for Long-Horizon Agents
Contrasts itself with data-centric MCP environment scaling by building from tools first.
NOW · Nov 5Agent-R² grounds synthesis in real-source data for productivity work1 source · show source ▾
Agent-R$^2$: Scaling Agents for Realistic Productivity Work via Real-Source Data Synthesis
Uses real-source data to synthesize training for realistic productivity work.
Day 1 Monday, Oct 19, 2026
Agent World Model in the Wild: Computer-Use Agent Environment Synthesis
Frames environment synthesis as a way to avoid the risks of learning directly in real computer systems.
Day 2 Tuesday, Oct 27, 2026
Scaling Laws for Tool-Use Agents via Controllable Environment Synthesis
Studies how data composition in synthetic tool-use environments affects agent performance.
Day 3 Tuesday, Nov 3, 2026
ToolSpace: Scaling Tool-Centric Environment Synthesis for Long-Horizon Agents
Contrasts itself with data-centric MCP environment scaling by building from tools first.
Day 4 Thursday, Nov 5, 2026
Agent-R$^2$: Scaling Agents for Realistic Productivity Work via Real-Source Data Synthesis
Uses real-source data to synthesize training for realistic productivity work.
What to watch — open questions
- Do any of these environment-synthesis methods report results replicated by an independent group?
- Are models trained on synthesized MCP environments released with weights or evals others can run?
- Does synthesized-environment training show gains on real-world productivity benchmarks, not only the authors' own?
How this thread was built
editor wrote the arc · 4 beatswatcher 1 status change
Storylines are threaded mechanically from the feed: stories that share a distinctive anchor across multiple days and sources. Each item links to its original source. The evidence trace, current state, and open questions are written by the editor routine and refreshed whenever a new beat lands.