JHU study finds in-context learning emerges across language, genome, protein and image models alike
DanielKhashabi · x · 2026-09-16
New research from JHU CLSP, "Convergent Emergence of In-Context Learning Across Modalities," shows that models trained on language, genome, protein, image, timeseries, and integer sequences all exhibit few-shot in-context learning.
Notably, their ICL performance is highly correlated across modalities, suggesting ICL is a convergent emergent phenomenon rather than something unique to language models.
Related event: JHU study finds in-context learning emerges across six modalities(4 posts)→
More from Research
- Xiaomi MiMo's CodeSpecWeave trains coding agents on interdependent specs for long-horizon coding — XiaomiMiMo · 2026-11-02
- Mistral Large 4 writes a quantum chemistry SCF solver from scratch with one short prompt — MistralAI · 2026-11-02
- Hunting for the most concrete "curing cancer with AI" roadmap of 2025 — iskander · 2026-11-02
- New theory paper proves linear decay is the unique minimax-optimal stepsize schedule for SGD — konstmish · 2026-11-02
- Karpathy's nanochat: the best ChatGPT $100 can buy, near 60k GitHub stars — karpathy · 2026-11-02
- After reading 46 resources on agent evals, one author explains why her 95% score was misleading — blaizedsouza · 2026-11-02