Xiaomi MiMo's CodeSpecWeave trains coding agents on interdependent specs for long-horizon coding
XiaomiMiMo · hf · 2026-11-02
Xiaomi's LLM Core team, with Peking University and University of Hong Kong, published "When Requirements Interact: Scaling Long-Horizon Coding via Interdependent Specifications."
- Problem: Real-world long-horizon software development requires coding agents to satisfy multi-dimensional, interdependent specifications (interface compatibility, error handling), which existing task-synthesis methods fail to capture.
- Method: CodeSpecWeave synthesizes complex software development tasks with interdependent specifications, paired with verified tests, to train coding agents.
- Results: Improved performance on long-horizon coding benchmarks.
A notable attempt to explicitly model spec interdependencies in training data construction for more capable coding agents.
More from coding & agent
- GitHub Actions has a lesser-known ubuntu-slim runner: 1 CPU, 5GB RAM, cheaper CI — DanielLockyer · 2026-11-02
- Before automating a workflow with AI, ask if anyone understands it end to end — djgoel123 · 2026-11-02
- One stale date silently broke a WhatsApp reply agent: 19 messages unanswered — petrusenko_max · 2026-11-02
- The 10X engineer is now just a SKILLS.md: 6 workflow rules for AI coding agents — mdancho84 · 2026-11-02
- Auditing my MCP credential broker: the worst hole was the config file — Relevant_Mousse_9117 · 2026-11-02
- AWS's open-source Strands Harness claims 45% cheaper than Claude Code, reigniting harness debate — blaizedsouza · 2026-11-02