Inference-Time Intervention: Eliciting Truthful Answers from a Language Model · HackerPapers discussions

Inference-Time Intervention: Eliciting Truthful Answers from a Language Model — mentioned by skhameneh

Sun, 03 Aug 2025 17:52:57 GMT

I was talking to an old colleague/friend about distillation, trying to understand how to steer distillation with regards to removing irrelevant regions of a larger model when training a smaller model. He shared this paper with me, calling the works seminal, it appears to be highly relevant: Inference-Time Intervention: Eliciting Truthful Answers from a Language Model https://arxiv.org/pdf/2306.03341

Paper and discussion history. A mention is not an endorsement.

Inference-Time Intervention: Eliciting Truthful Answers from a Language Model — mentioned by agucova

Sun, 21 Jul 2024 15:49:03 GMT

> If training data contains multiple conflicting perspectives on a topic, the LLM has a limited ability to recognize that a disagreement is present and what types of entities are more likely to adopt which side. That is what those studies are reflecting. Again, we have empirical evidence to suggest otherwise. It's not that there's an oracle, but that the LLM does internally differentiate between facts it has stored as simple truth vs. misconceptions vs. fiction. This becomes obvious by interacting with popular LLMs; they can produce decent essays explaining different perspectives on various…

Paper and discussion history. A mention is not an endorsement.

Inference-Time Intervention: Eliciting Truthful Answers from a Language Model — mentioned by astrange

Wed, 24 Apr 2024 05:04:03 GMT

> demonstrate that the LLM will deny that they know fact X (or be flaky about it, randomly denying and divulging the fact) No, the sampling algorithm you used to query the LLM does that. Not the model itself. e.g. https://arxiv.org/pdf/2306.03341.pdf > B. They know they know fact X, but avoid acknowledging for ... reasons? That reason being that the sampling algorithm didn't successfully sample the answer.

Paper and discussion history. A mention is not an endorsement.

Inference-Time Intervention: Eliciting Truthful Answers from a Language Model — HN submission

Fri, 30 Jun 2023 16:09:49 GMT

Inference-Time Intervention: Eliciting Truthful Answers from a Language Model

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Inference-Time Intervention: Eliciting Truthful Answers from a Language Model — mentioned by cubefox

Tue, 13 Jun 2023 16:07:14 GMT

Yeah. See e.g. https://arxiv.org/abs/2306.03341 > Our findings suggest that LLMs may have an internal representation of the likelihood of something being true, even as they produce falsehoods on the surface. The problem here is that there is currently no reliable way to extract information from this hypothetical world model. Language models do not always say what they "believe", they might instead say what is politically correct, what sounds good etc. Researchers try to optimize (fine-tune) language models to be helpful, honest, and harmless, but honesty ("truthfulness") can't be easily…

Paper and discussion history. A mention is not an endorsement.

Inference-Time Intervention: Eliciting Truthful Answers from a Language Model — HN submission

Thu, 08 Jun 2023 19:29:21 GMT

Inference-Time Intervention: Eliciting Truthful Answers from a Language Model

4 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.