TDmol uses 2D molecules as a bridge: text guidance boosts 3D structure similarity by 41%
bravo_abad · x · 2026-09-03
TDmol sidesteps scarce text-3D pairing data by using 2D molecular structure as a bridge: learn text↔2D, then 2D↔3D, and use those aligned representations to guide an equivariant diffusion model building molecules atom by atom. Across 234,400 generated molecules, adding text guidance increased similarity between the requested description and generated 3D structure by 41%—enabling natural-language-to-3D molecule generation.
More from Research
- Startup Mostik bridges AI models via their weights, tops ARC-AGI 3 at 1/20 the cost — nordicinst · 2026-09-03
- ByteDance's looped language models match 12B rivals at 1.4B size, with Bengio as co-author — peterjliu · 2026-09-03
- OpenAI's CoT monitorability hit: paper authors double down on 'fragile' AI safety window — GaryMarcus · 2026-09-03
- PufferLib author says retuning brings ~3x end-to-end speedup, up to 10x in some envs — yacineMTB · 2026-09-03
- Harvard Proposes Agentic Data Cracking, Cuts Unstructured QA Cost 53% on FanOutQA — Harvard · 2026-09-03
- Student runs explainable bone-lesion X-ray screener for £5/month — xrY- · 2026-09-03