New weekly newsletter filters AI-for-Science papers down to 3-5 transferable ideas
bravo_abad · x · 2026-09-18
Physicist Jorge Bravo-Abad runs Discovery at Scale, a free weekly Substack that filters hundreds of weekly AI-for-Science papers down to 3-5 ideas researchers can actually carry into their own work. Each issue groups findings by what they teach rather than paper by paper, and includes: the concrete ML mechanism behind each result, one number worth remembering, a clear split between what is demonstrated versus speculative, and the lesson that transfers beyond the original field. Aimed at researchers in physics, chemistry, materials science, biology, math and ML.
More from Research
- Paper: Suppressing LLM self-attributions shifts reported values away from human norms — coherence · 2026-09-18
- Amazon's ActObs: supervising observation tokens in SFT boosts agent RL exploration and pass@k — amazon · 2026-09-18
- New Multiscale Emergence Methods Applied to EEG Across Conscious States, Paper Published — anilkseth · 2026-09-18
- AndroidLife real-phone benchmark: Qwen3.8-27B fails 43% of 60 daily tasks — East-Muffin-6472 · 2026-09-18
- WeirdML v3 launches: agentic benchmark with 11 hand-made ML tasks — scaling01 · 2026-09-18
- Stanford researchers launch CoPaper, an AI co-authoring platform already behind published journal papers — JeremyNguyenPhD · 2026-09-18