How does "outperform" translate to the propensity of an LLM to hallucinate?
There seems to be a mass delusion about how capable SOTA models actually are. That's my only explanation for how poorly I find them performing in basic knowledge tasks compared to how others describe their prowess.
alt Hacker News
There seems to be a mass delusion about how capable SOTA models actually are. That's my only explanation for how poorly I find them performing in basic knowledge tasks compared to how others describe their prowess.