Pre-LLM NLP predicts substance behind de-identified psychedelic trip reports far above chance
Josikinz · x · 2026-10-03
Josikinz reports that a semantic model built with old-school pre-LLM NLP predicted the substance behind de-identified psychedelic trip reports an order of magnitude better than random chance. Misclassifications were also non-random: subjectively similar drugs such as MDA were commonly mistaken for MDMA.
Related event: Pre-LLM NLP Models Can Identify Psychedelics from Trip Reports(2 posts)→
More from Research
- Using an LLM to Derive Interpretable Labels for PLSR Components in Drug Effect Space — Josikinz · 2026-10-03
- NSF Renews AI Institute for Engaged Learning for 5 Years, Betting on 3D World Models — deliprao · 2026-10-03
- RSI Arena live-streams AI agents training models, judged by human evaluation — kuchaev · 2026-10-03
- BIND Action Head Ties Robot Actions to 2D Image Features for Data-Efficient Policies — yuewang314 · 2026-10-03
- JEPA-TTT: Persistent test-time training lifts world model planning by 153% — DanielKhashabi · 2026-10-03
- ByteDance's DMAD distills MiniMax H3 video to 4 steps — AgeNo5351 · 2026-10-03