Andrej Karpathy

LLMs with memory/personalization features tend to over-index on past user queries, repeatedly surfacing old topics as if they represent deep interests, resulting in irrelevant and intrusive mentions that feel forced rather than helpful.

One common issue with personalization in all LLMs is how distracting memory seems to be for the models. A single question from 2 months ago about some topic can keep coming up as some kind of a deep interest of mine with undue mentions in perpetuity. Some kind of trying too hard.

<p>One common issue with personalization in all LLMs is how distracting memory seems to be for the models. A single question from 2 months ago about some topic can keep coming up as some kind of a deep interest of mine with undue mentions in perpetuity. Some kind of trying too hard.</p>