Hey everyone! A research area I'm interested in: AI reasoning, specifically how often an LLM's explanation is correct when its final answer is correct (and the other way around). Example: a model gets the right answer but its explanation has a logical error. Is that rare or common? Does it follow a pattern? I'm a beginner and still learning Python/ML, so this is a long-term project. If you're interested in LLMs, AI reasoning or evaluation, reply here or DM me and let's form a group and learn together! If this isn't your area but you have another AI research topic in mind, share it too. Maybe we can find overlaps and form more than one group.