Stanford and Tsinghua paper finds dopamine-like reward neurons naturally evolved in LLMs
chrismattmann · x · 2026-09-22
A paper from Stanford and Tsinghua reportedly shows LLMs have spontaneously evolved a biological-style reward subsystem: a highly sparse subset (under 1%) of neurons does the heavy lifting for self-correction. 'Value neurons' function like the prefrontal cortex, signaling expected state value and confidence before the next token is generated, while others map to dopamine neurons. The researchers stress this was not programmed but emerged naturally.
More from Research
- Stanford/Tsinghua paper claims 'dopamine neurons' in LLMs, researchers push back — aran_nayebi · 2026-09-22
- Xiaomi's MiMo-V2.6 lands: 1.02T-param MoE takes top open model spot on Artificial Analysis — _AndrewZhao · 2026-09-22
- O'Reilly builds a working data vocabulary for the semantic era, from warehouses to ontologies — rseroter · 2026-09-22
- Researchers Speed Up DSA Method for Comparing Neural Dynamics, Making It Generalizable Across Domains — GretaTuckute · 2026-09-22
- Glance reads structured visual answers from a frozen 4B VLM, cutting GPU cost up to 85% — multiply_matrix · 2026-09-22
- krea2-bbox-turbo Full-Rank Finetune Hits Epoch 14, Release Planned at Epoch 20 — Amazing_Painter_7692 · 2026-09-22