๐ค Things I am currently curious: how to combine the reinforcement learning (search) and (self-) supervised learning for building general multimodal agents
โก Fun fact: I used to be a good swimmer who grasp 4 sytles including Butterfly, Backstroke, Breaststroke and Freestyle. However, because of staying up late for a long time, I can no longer swim very far in a row. Take me as a warning to protect your body! (And if you are in HK, I'm very willing to teach you my skills!
Hello, thanks for providing this awesome repository introducing different chain-of-thought variants!
Could you consider adding our CoT Collection dataset? It's a massive instruction dataset consisted of 1.84 million rationales across 1060 NLP tasks for CoT Fine-tuning!