PaulScotti

7 ms·

1. 1.▲
   Why open multimodal medical foundation models will be the future of healthcare(paulscotti.substack.com)

   4 points by PaulScotti 2y ago|0 comments
2. 2.▲
   by PaulScotti 2y ago

   Thanks for sharing! I read your README and think it's a very interesting research path to consider. I wonder if such an adversarial game approach could be outfitted to not just well-defined games but to wholly generalizable improvement
3. 3.▲
   by PaulScotti 2y ago

   To anyone still finding the game slow due to traffic, you can just git clone the game, add your ANTHROPIC API key to a .env file, and play it locally (this is explained in the README in our github repo). It runs super fast if played locally
4. 4.▲
   by PaulScotti 2y ago

   Thanks! Yes the traffic is making the game slow... To anyone impatient you can just git clone the game, add your ANTHROPIC API key to a .env file, and play it locally. It runs super fast if played locally. Creating a service would be amazin
5. 5.▲
   by PaulScotti 2y ago

   Yeah sorry, it is still quite slow due to the traffic. It'd be much faster and robust to run locally via git cloning the repo and adding your own API key as shown in the README For using other models it should be pretty straightforward
6. 6.▲
   by PaulScotti 2y ago

   Yeah nice idea -- does sound plausible and would make things much cheaper and faster.
7. 7.▲
   by PaulScotti 2y ago

   The Officer doesn't actually get supplied information on the true killer in their context window... so that response you got is actually incorrect. You can check the actual solution by clicking the End Game button
8. 8.▲
   by PaulScotti 2y ago

   Upgraded the server and should now be working... I think
9. 9.▲
   by PaulScotti 2y ago

   Should be working now and way faster! Had to upgrade the server to increased number of workers
10. 10.▲
    by PaulScotti 2y ago

    Haha interesting approach!
11. 11.▲
    by PaulScotti 2y ago

    Damn that sucks, sorry. For what it's worth I tried playing the game dozens always asking for an overview as my first message and I never encountered such a response , so hopefully that's quite the rare experience.
12. 12.▲
    by PaulScotti 2y ago

    Wow really, can you tell me what you said to get them to confess?
13. 13.▲
    Multi-agent chatbot murder mystery(ai-murder-mystery.onrender.com)

    302 points by PaulScotti 2y ago|128 comments
14. 14.▲
    by PaulScotti 2y ago

    Sharing a little open-source game where you interrogate suspects in an AI murder mystery. As long as it doesn't cost me too much from the Anthropic API I'm happy to host it for free (no account needed). The game involves chatting
15. 15.▲
    by PaulScotti 3y ago

    Guys Figure 1 is not real results, it's an illustration of the "goal" of the paper. The real results are in Table 3. And are much worse.
16. 16.▲
    by PaulScotti 4y ago

    There's already been preprints released showing fMRI reconstructions that appear to do better than an implicit multi-class classifier [1] [2]. But also, even if the result is an implicit multi-class classifier, if the n is sufficiently
17. 17.▲
    by PaulScotti 4y ago

    Exciting stuff, looking forward to seeing how this current AI revolution leads to improvements in the medical research space -- seems very promising