idea
- Improving āLLM as a judgeā schemes by ensuring minimal semantic correlation between evaluation dimensions
- āLLMās donāt want to do workā - input a stipulation where an LLM is rating something that is supposed to be neutral, but stipulate that when it decides to give a bad rating it has to explain more
- first would have to research bias of LLMs in rating tasks in general.
- Align a model to ask questions more often, āask and you shall receive"
- "Entering the flow state with LLMsā, use the sparse autoencoders idea where you can add a vector at the last layer and find out how much that improves LLM output
- For robots arms that use cameras that are directly on the grippers, how can we introduce behaviour to ātake a step back and look at the bigger pictureā, before going back in?