LLMs of course, but think the real issue here is incentives. Clearly the participants are incentivized to publish garbage. That's going to crater the signal to

LLMs of course, but think the real issue here is incentives. Clearly the participants are incentivized to publish garbage. That's going to crater the signal to noise ratio of papers so best that be fixed asap. How though...