3 ms·
There's Lots of evidence indicating language models roughly know when they hallucinate. GPT-4 logits calibration pre RLHF - [https://imgur.com/a/3gYel9r](https
by famouswaffles 3y ago
There's Lots of evidence indicating language models roughly know when they hallucinate.
GPT-4 logits calibration pre RLHF - [https://imgur.com/a/3gYel9r](https://imgur.com/a/3gYel9r) https://imgur.com/a/3gYel9r](https://imgur.com/a/3gYel9r)
Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback - [https://arxiv.org/abs/2305.14975](https://arxiv.org/abs/2305 https://arxiv.org/abs/2305.14975](https://arxiv.org/abs/2305...
Teaching Models to Express Their Uncertainty in Words - [https://arxiv.org/abs/2205.14334](https://arxiv.org/abs/2205 https://arxiv.org/abs/2205.14334](https://arxiv.org/abs/2205...
Language Models (Mostly) Know What They Know - [https://arxiv.org/abs/2207.05221](https://arxiv.org/abs/2207 https://arxiv.org/abs/2207.05221](https://arxiv.org/abs/2207...
It seems very much like hallucinations aren't a randomness or representation issue. The computation already knows. It just doesn't care about telling you this
- corobo 3y agoI've been using it as a joke but if they're predicting what we want to read, is it possible they just learned to make stuff up by having internet forums in their training data? The amount of BS we feed each other online, maybe it just thinks we'd prefer something made up than "I don't know"
- corobo 3y agoNot sure if it came off as me saying the parent comment was BS or if what I said was just dumb. Downvotes don't explain anything, haha For the record I wasn't saying the parent comment was BS :)
- QuadmasterXLII 3y agoRLHF also drops GPT3.5’s chess elo from ~1800 to ~800 which i confabulate to be connected
- naveen99 3y agoIs this gpt3.5 Elo a modern chess.com elo ? I feel like there has been some deflation in elo in the last 2 years.
- anon291 3y ago> It just doesn't care Matrix multiplications cannot feel anything so of course it doesn't care. Humans care. Language models do not.
- hfhdjdks 3y agoAlso 2 == 2. You're not making an argument, you're just saying that LLMs aren't human.