[HN Gopher] LLM Hallucinations: why they occur and how to avoid ...
       ___________________________________________________________________
        
       LLM Hallucinations: why they occur and how to avoid them
        
       Author : ofermend
       Score  : 19 points
       Date   : 2023-05-02 20:50 UTC (2 hours ago)
        
 (HTM) web link (vectara.com)
 (TXT) w3m dump (vectara.com)
        
       | yawnxyz wrote:
       | That's a lot of words to say "we feed it up to date SERP results"
        
       | mdp2021 wrote:
       | > _When the language model's predictions contradict our
       | expectations, experiences or prior knowledge, or when we find
       | counter-factual evidence to that response (sequence of predicted
       | tokens) - that's when we find hallucinations_
       | 
       | No. Hallucination is any idea which was not assessed for truth.
       | Whatever statement is not put over the "testing table" and
       | analyzed foundationally counts as hallucination.
        
       | valine wrote:
       | The proposed solution to eliminate hallucination is to ground the
       | model with external data. This is the approach taken with Bing
       | Chat, and while it kinda works, it doesn't play to the strengths
       | of LLMs. Every time Bing Chat searches something for me I can't
       | help but feel like I could have written a better search query
       | myself. It feels like a clumsy summarization wrapper around
       | traditional search, not a revolutionary new way of parsing
       | information.
       | 
       | Conversing with an LLM on subjects that it's well trained on,
       | however, absolutely does feel like a revolutionary new way of
       | parsing information. In my opinion we should be researching ways
       | to fix hallucinations in the base model, not papering over it by
       | augmenting the context window.
        
         | og_kalu wrote:
         | Fixing hallucinations from the source will be a tough one. The
         | root of the issue is that loss doesn't discriminate. A probable
         | guess will lower loss much better than "I don't know" or
         | whatever equivalent. Educated guessing becomes an essential
         | skill the model learns during initial training.
         | 
         | So the objective function encourages it. But also the dataset
         | encourages it as well. There will be many many sentences that
         | can't be completed accurate to source even with all the
         | knowledge and understanding in the world. Many completions will
         | have numerous sensible options. The dataset doesn't
         | discriminate. Fiction, Fact, Opinion. All the same. All given
         | equal weight.
        
         | roywiggins wrote:
         | Even Bing AI seems to sometimes get confused and fail to
         | summarize results properly, as when seems to have parsed a
         | debunking of a ChatGPT hallucination but still spit out the
         | hallucination again:
         | 
         | https://twitter.com/WillOremus/status/1643692259332743171
        
         | electric_mayhem wrote:
         | Tl;dr: I think that a left brain/right brain parallel is in the
         | cards for LLMs
         | 
         | Our own brains have multiple neural circuits. Parallel, serial,
         | competing, cumulative...
         | 
         | Including ones which error-check others' output.
         | 
         | I guess the term in the ML niche is 'adversarial networks'.
         | 
         | From the robotics side, there was subsumption architecture
         | which used the real world as a basis for informing decisions.
         | 
         | So, I respectfully disagree that fixing up creative but
         | occasionally erroneous networks is papering over the problem.
         | 
         | If our own brains use multiple neural circuits to check each
         | other to try to optimize for output that agrees with reality,
         | why would it be preferable or logical to try to create a single
         | artificial neural network that has perfect output?
         | 
         | My sense from playing with LLMs is they're knowledge without
         | understanding. Weirdly akin to my own dreaming consciousness,
         | to the extent that conscious me has seen it.
         | 
         | It seems very natural to me that the next step would be
         | integrating checks/validations in parallel and considering each
         | half of a matched 'whole'.
         | 
         | It could still be considered a single network. Just made of two
         | distinct smaller specialized ones.
        
           | valine wrote:
           | Having a second adversarial network could absolutely be part
           | of the solution, I wasn't arguing against that. My problem is
           | with in context learning, ie the idea that including
           | background information in the prompt can solve hallucination.
        
             | electric_mayhem wrote:
             | I see where you're coming from, but I've had a couple
             | occasions over the years that have led me to expect and
             | accept that at any given time my brain harbors an abundance
             | of different ideas and not all of them are valid.
             | 
             | Like the time I decided to sit on the ceiling to spite a
             | friend who was being pushy.
             | 
             | Made perfect sense, and I was about to. Until a different
             | part kicked in and assured the whole of myself that it
             | wasn't possible.
             | 
             | Neural networks are going to spawn occasional nonsense.
             | Aiming for perfection is noble, but nature shows a
             | different path that already works.
             | 
             | But fundamentally I think we're agreeing- just the twist is
             | I'm saying LLMs should be an aspect of a whole rather than
             | seen as a solitary solution to be perfected.
        
         | cyanydeez wrote:
         | The problem is you're proposing, basically to implement a
         | "ASKJEEVES" model for Wikipedia, with a unknown concentration.
         | 
         | That'd cost real money, requires real licensing, and a clear
         | business plan.
         | 
         | Absolutely it'll happen, but almost every sector will need a
         | leader for X subject.
        
           | valine wrote:
           | Relying on subject matter experts isn't a scalable solution.
           | I would speculate that LLMs don't hallucinate when a fact is
           | included a handful of times in the training set. I'd also
           | speculate that our current LLMs can say whether a piece of
           | text looks like a fact.
           | 
           | I imagine the long term solution to hallucination will look
           | something like this: Loop over the entire training set with
           | the latest GPT and build a list of "facts", then train a new
           | LLM using this new dataset to predict both the probability of
           | the next token, and the probability that the next token is
           | factual.
        
       | redskyluan wrote:
       | I would vote for finetuning, prompt engineering, rather than only
       | add domain specific knowledge. Others are playing detective
       | ,digging into the ethical conundrums of AI-generated content.
        
       | eskibars wrote:
       | I know the article doesn't go into this particular element, but I
       | do wonder how much opportunity is still in front of us for
       | adversarial LLM systems that try to detect/control for
       | hallucinations. I'm pretty excited by the research in LLM
       | explainability and quantitative measures on how accurate
       | generative LLMs are measured
       | 
       | (Full disclosure: I work at Vectara, where this blog was
       | published)
        
         | cornhole34 wrote:
         | Please do not anthropomorphize LLM.
        
           | eskibars wrote:
           | I wasn't, and I'm not sure how you got that out of what I
           | said. I'm not claiming "understanding," "sentience," etc.
           | 
           | I'm claiming there's a great deal of work in the realm of
           | research that I'm excited by: research I expect to be done by
           | humans. I do expect/am supposing that result may be LLMs of a
           | different nature: to apply guard rails around generative LLM
           | systems, but that's not to anthropomorphize them: just to
           | suppose their purpose.
           | 
           | The term "hallucination" does anthropomorphize LLMs, but I
           | think that's now accepted nomenclature in the industry, at
           | least for the time being, and it's helpful to have some
           | standard nomenclature to describe some of the benefits and
           | problems.
        
       ___________________________________________________________________
       (page generated 2023-05-02 23:01 UTC)