Packora debuts: a generative model for molecular crystal structure prediction
CatAstro_Piyush · x · 2026-09-12
Researchers introduced Packora, a state-of-the-art generative model for molecular crystal structure prediction (CSP). Key features: flexible optional conditioning, a two-track evaluation with leading performance, and extensive systematic ablations. The work applies generative modeling to crystal structure prediction in materials science.
More from Research
- New paper: self-replicating programs evolve cooperation under a shared energy budget — irinarish · 2026-09-12
- CodeFinetuner: fine-tune a local code autocomplete model on your own codebase — MountainTop321 · 2026-09-12
- 25 Fields Medallists including Terence Tao declare severe misalignment of AI in mathematics — natolambert · 2026-09-12
- Matei Zaharia's team relaunches large-scale survey of production AI agents — matei_zaharia · 2026-09-12
- Persistent-memory denoiser hits 99.9% on Sudoku-Extreme with under 250k parameters — LucaAmb · 2026-09-12
- MIT's Astra invents its own scientific instruments, finds metamaterial 2.1x stronger than reference — ProfBuehlerMIT · 2026-09-12