Tsinghua and Peking University Probe Whether LLMs Think Beyond Language in Nature MI
jiqizhixin · x · 2026-10-04
Neuroscience shows the brain's language network is separable from reasoning systems — language can be impaired while complex reasoning persists. Yet LLMs, trained purely via next-token prediction, already reason, plan, and use tools. Tsinghua and Peking University researchers address this gap in Nature Machine Intelligence, asking whether LLMs merely exploit statistical patterns or have formed internal representations resembling human cognition. Building on prior work linking LLM representations to brain language-network activity, the paper examines logical reasoning as a higher-order cognitive task.
More from Research
- A gentle introduction to phase-locked loops, the Kuramoto model, and gyrotrons — stevenstrogatz · 2026-10-04
- ICLR's official .bib template has listed Bengio twice since 2019 — tughanbulut · 2026-10-04
- kalomaze: In LoRA, Weight Decay Behaves Like Regularizing Toward the Base Model — kalomaze · 2026-10-04
- Humans are vastly more sample-efficient than LLMs, and finding why could rival attention — burny_tech · 2026-10-04
- Judea Pearl Announces Second Edition of The Book of Why, Due October 20 — yudapearl · 2026-10-04
- Nature study: policies that spread cooperation concentrate benefits among the well-connected — arjunrajlab · 2026-10-04