ChefNMR2 released: generates 3D molecular structures from 1D NMR with calibrated uncertainty
ZhongingAlong · x · 2026-11-03
The team officially releases ChefNMR2, which takes 1D proton/carbon NMR peak lists plus a molecular formula (e.g. from mass spec) and generates candidate molecular structures directly.
- Upgraded from diffusion to flow-matching, with a new architecture based on multimodal diffusion transformers.
- Training data massively expanded to 2.52 million molecules spanning experimental spectra, synthetic spectra, patented molecules, and complex natural products.
- The model generates atomic coordinates directly (akin to protein structure modeling) rather than SMILES or graphs, targeting large intricate natural product structures.
- Outputs come with calibrated uncertainties that can guide interpretation.
More from Research
- Researcher urges critics to read the prompts in the 'weight-space programming' post first — anton_lozhkov · 2026-11-03
- New paper: standard CoT monitorability metric is only an upper bound, improvable via self-distillation and RL — armancohan · 2026-11-03
- New Scaling Recipe for Data-Constrained Training Cuts Perplexity 14% via Hyperparameter Transfer and Ensembling — andrewgwils · 2026-11-03
- Purdue professor says arXiv banned him a year over 'hallucinated references' that weren't — RexDouglass · 2026-11-03
- KV cache translator proposal: small model reads 100K tokens, large model corrects — dreamai87 · 2026-11-03
- Yoav Goldberg highlights InterpBench, a benchmark of Transformers with known circuits for interpretability — yoavgo · 2026-11-03