EPFL's GOLLuM turns AI uncertainty into a training signal, cutting lab trials by 40%
MacrinePhD · x · 2026-09-03
Researchers at EPFL built GOLLuM, a framework combining LLMs with Gaussian processes to turn uncertainty into a training signal. By learning from 'doubt,' it maps experimental recipes faster and matches traditional methods using 40% fewer lab trials, targeting AI-optimized scientific discovery.
Related event: EPFL's GOLLuM Teaches AI to Learn from Doubt(2 posts)→
More from Research
- TrackEverything: CMU and Meta open-source a 3D tracker for all points in 1000+ frame videos — AdamWHarley · 2026-10-21
- Infantconnect: open-source infant brain anatomy and EEG connectivity tool from 1 to 72 months — Dr_Alex_Crimi · 2026-10-21
- Berkeley wins Schmidt Sciences HAVI grant to decode whale and parrot communication with AI — begusgasper · 2026-10-21
- Stanford CS336 Lecture 6: fused Triton kernels cut softmax memory traffic ~4x — stanfordnlp · 2026-10-21
- NVIDIA Cosmos3-H-Dreams: Generative Simulation Brings Clinical Surgery to Life — Hugging Face Blog · 2026-10-21
- UIUC Paper: High Accuracy Hides Flawed Reasoning as LLMs Fail Under Rephrased Logic Problems — UIUC-CS · 2026-10-21