GBrain adds brainstorm and LSD modes that apply embeddings to surface contrarian ideas from user notes, citing OpenCollider research and benchmarks of 49.1% P@5 and 97.9% R@5 on a 240-page corpus
Garry Tan posted command examples for the new modes.
I find some of the most fun and thought provoking ideas do come from connecting distant domains
This is one of the most fun features I've built so far for my OpenClaw setup of now 120k markdown files
Garry Tan@garrytanGBrain now can use the embeddings to find you contrarian ideas using 'brainstorm' and 'brainstorm with lsd mode (lateral synaptic drift)' Researchers at OpenCollider showed context eng techniques can boost originality Now you can try it with your own ideas in your brain
GBrain is my brain and infra and skillpacks, all open sourced for your use, MIT license
It's SOTA for LongMemEval without LLM query rewriting by 1% vs MemPalace
If you use OpenClaw or Hermes Agent this is what you should use to supercharge your agent
Garry Tan@garrytanGBrain now can use the embeddings to find you contrarian ideas using 'brainstorm' and 'brainstorm with lsd mode (lateral synaptic drift)' Researchers at OpenCollider showed context eng techniques can boost originality Now you can try it with your own ideas in your brain
#270Garry Tan@GARRYTANIdea fusion is fun, and idea fusion with LSD (lateral synaptic drift) is even more fun
#1894Nick Dobos@NICKADOBOSGarry gave his AI agent LSD
Genuine question, can we do this for all drugs & mental illnesses? What does that look like?
What does giving a folder of memory .md files ADHD or cocaine actually mean?
How would each one work? Can we model that?
I’m very curious what patterns show up and what that means for how we understand human thinking
Garry Tan@garrytanGBrain now can use the embeddings to find you contrarian ideas using 'brainstorm' and 'brainstorm with lsd mode (lateral synaptic drift)' Researchers at OpenCollider showed context eng techniques can boost originality Now you can try it with your own ideas in your brain