Firstly, many thanks for your incredible work, and congrats on achieving such great results on a difficult task like ARC! I wanted to ask if you are comfortable sharing the results you obtained on the public test set, i.e. the output images from the language model, such that it saves others the time of running your experiment? If you can share the entire conversation (in-context examples, chain of thought, LLM output etc) that would be even better, because replicating your experiments will consume significant time and resources. Thanks again for the great work!
Dear Ryan,
Firstly, many thanks for your incredible work, and congrats on achieving such great results on a difficult task like ARC! I wanted to ask if you are comfortable sharing the results you obtained on the public test set, i.e. the output images from the language model, such that it saves others the time of running your experiment? If you can share the entire conversation (in-context examples, chain of thought, LLM output etc) that would be even better, because replicating your experiments will consume significant time and resources. Thanks again for the great work!