A: Well I appreciate you saying so Greg!! Means a lot! We work hard ;) up til now I’ve only heard of CoT prompting of chain of thought. The YouTube video you posted is only 6 hours old lol. So not sure how new that tech is, but it seems far off. Given this is a white-paper and it’s showing how VoT can be done with LLMs, but it’s not currently, shows me it’s very new and the most cutting edge LLMs aren’t adopting it into their algos. Even at one point the YouTuber here quoted someone with research that says it may not even be possible with LLMs. This is interesting but it seems more science fiction than anything else. For a machine to have a “minds eye” in the next 5 years idk if that’s going to happen. Right now they hallucinate 1/3-1/2 of the time and can’t be trusted on their own. If this happens, great for LLMs but then how proprietary is the data that machine is trained on? It’s more about the data sets; bc all LLMs will become ubiquitous in the next 5 years or so. Taskmagic will be best in class with proprietary data to train for our customers ;) thanks for sharing this btw