Banishing LLM Hallucinations Requires Rethinking Generalization · HackerPapers discussions

Banishing LLM Hallucinations Requires Rethinking Generalization — mentioned by nickpsecurity

Mon, 27 Jan 2025 21:51:28 GMT

I forgot one tech that’s actually commercialized by lamini.ai: https://arxiv.org/abs/2406.17642 https://medium.com/pythons-gurus/mixture-of-memory-experts-l... Re your reply I’ll add two things to “drive us to models that have emergent, reasoning abilities:” 1. It didn’t have that until they spent around $30 million pre-training a specific architecture (GPT3-176B) on TB’s of data with portions having built-in, reasoning steps it could extrapolate. Doing the same with the other architectures might have led to similar, better, or worse results. GPT3’s training investment, not Transformer…

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

Banishing LLM Hallucinations Requires Rethinking Generalization — mentioned by Animats

Fri, 15 Nov 2024 07:27:14 GMT

Comments on that paper? PDF: [1] What they are measuring, it seems, is whether LLMs can be built which will retrieve a reliable known correct answer on request. That's an information retrieval problem, and, in fact, they solve it by adding "Memory Experts" which are basically data storage. It's not clear that this helps either replies which require synthesizing disparate information, or detecting that the training data does not contain info needed to construct a reply. [1] https://arxiv.org/pdf/2406.17642

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

Banishing LLM Hallucinations Requires Rethinking Generalization — mentioned by nickpsecurity

Fri, 15 Nov 2024 04:25:00 GMT

The brain solves that problem. It seems to involve memory and specialized regions. I found a few groups building hippocampus-like, research models. One had content-addressable memory. There was another one that claimed to get rid of hallucinations. They also said it takes 50-100 epochs for regular architectures to actually memorize something. Their paper is below in case people qualified to review it want to. https://arxiv.org/abs/2406.17642 Like the brain, I believe the problem will be solved by a mix of specialized components working together. One of those components will be a memory (or…

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

Banishing LLM Hallucinations Requires Rethinking Generalization — mentioned by gdiamos

Sat, 14 Sep 2024 18:39:25 GMT

Disagree - https://arxiv.org/abs/2406.17642 We cover halting problem and intractable problems in the related work. Of course LLMs cannot give answers to intractable problems. I also don’t see why you should call an answer of “I cannot compute that” to a halting problem question a hallucination.

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

Banishing LLM Hallucinations Requires Rethinking Generalization — mentioned by gdiamos

Tue, 27 Aug 2024 19:49:45 GMT

We know that LLMs hallucinate, but we can also remove them. I’d love to see a future generation of a model that doesn’t hallucinate on key facts that are peer and expert reviewed. Like the Wikipedia of LLMs https://arxiv.org/pdf/2406.17642 That’s a paper we wrote digging into why LLMs hallucinate and how to fix it. It turns out to be a technical problem with how the LLM is trained.

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