Advertisement · 728 × 90

Posts by Victor Veitch

come learn about LLM geometry!

11 months ago 1 0 0 0

I'll present this poster tonight at East exhibit hall a-c 2510. 5-7:30 pm.

Come chat about alignment!

1 year ago 7 0 0 0

I'll be at NeurIPS Thursday-Sunday; send me an email if you'd like to chat :)

1 year ago 2 0 0 0
On Spurious Associations and LLM Alignment Large language models are `aligned' to bias them towards outputting responses that are good on various measures---e.g., we may want them to be helpful, factual, and polite. Often, alignment procedures...

in talk form simons.berkeley.edu/talks/victor...

1 year ago 4 0 0 0
On Spurious Associations and LLM Alignment Large language models are `aligned' to bias them towards outputting responses that are good on various measures---e.g., we may want them to be helpful, factual, and polite. Often, alignment procedures...

LLM Alignment aims at making model outputs preferred by a ranker while changing as little 'off-target' behavior as possible.

Turns out:
-best-of-$n$ is the optimal option!
-you can contrastively train an LLM to mimic its own best-of-$n$ distribution!

BonBon alignment: arxiv.org/abs/2406.00832

1 year ago 6 0 1 1