FrontierSmith lands NeurIPS spotlight: AI-synthesized coding data beats expert curation
AccBalanced · x · 2026-09-25
The FrontierCS team announces FrontierSmith, a system for synthesizing open-ended coding training data at scale, has been accepted to NeurIPS as a spotlight. Starting from closed-ended tasks, it mutates, filters, and builds runnable optimization environments for long-horizon coding agents. Models trained on its data outperform human-expert-curated data on FrontierCS and ALE-bench, suggesting open-ended coding data is no longer the bottleneck. Paper, code, and model are open-sourced.
More from coding & agent
- Clinician-Reviewed Child Psychiatry MCP Library Launches on Glama — modelcontextprotocol · 2026-09-25
- Brainiall MCP Server Brings Phoneme-Level Speech Assessment to Agents — modelcontextprotocol · 2026-09-25
- Plan Mode rework: customizable prompt, shareable modes, rebindable Shift+Tab — trq212 · 2026-09-25
- Firecrawl raises $75M Series B, targets $100M ARR with fewer than 50 people — devdigest · 2026-09-25
- Hill Sampling: Condition on the Best Verified Solution, Generate 512 Parallel Edits, Repeat — LChoshen · 2026-09-25
- Free Public MCP Server Exposes Canadian Privacy Law Data: 263 Terms, CAI Cases — masiha97 · 2026-09-25