[HN Gopher] Explaining Large Language Models Decisions Using Sha...
       ___________________________________________________________________
        
       Explaining Large Language Models Decisions Using Shapley Values
        
       Author : veryluckyxyz
       Score  : 80 points
       Date   : 2024-12-28 00:44 UTC (22 hours ago)
        
 (HTM) web link (arxiv.org)
 (TXT) w3m dump (arxiv.org)
        
       | goldemerald wrote:
       | While I love XAI and am always happy to see more work in this
       | area, I wonder if other people use the same heuristics as me when
       | judging a random arxiv link. This paper has one author, was not
       | written in latex, and no comment referencing a peer reviewed
       | venue. Do other people in this field look at these same signals
       | and pre-judge the paper negatively?
       | 
       | I did attempt to check my bias and skim the paper, it does seem
       | well written and takes a decent shot towards understanding LLMs.
       | However, I am not a fan of black-box explanations, so I didn't
       | read much (I really like Sparse autoencoders). Has anyone else
       | read the paper? How is the quality?
        
         | refulgentis wrote:
         | > I wonder if other people use the same heuristics as me when
         | judging a random arxiv link.
         | 
         | My prior after the header was the same as yours. The fight and
         | interesting part is in the work past the initial reaction.
         | 
         | i.e. if I react with my first order, least effort, reaction,
         | your comment leaves the reader with a brief, shocked, laugh at
         | you seemingly doing performance art. A seemingly bland
         | assessment and overly broad question...only to conclude with
         | "Has anyone else read the paper? Do you like it?"
         | 
         | But that's not what you meant. You're geniunely curious if its
         | a long tail, inappropriate, reaction to have that initial
         | assessment based on pattern matching. And you didn't mean "did
         | anyone else read it", you meant "Humbly, I'm admitting I'm
         | skimmed, but I wasn't blown away for reasons X, Y, and Z. What
         | do you all think? :)"
         | 
         | The paper is superb and one of the best I recall reading in
         | recent memory.
         | 
         | It's a much whiter box than Spare Autoencoders. Handwaving what
         | a bag of floats _might_ do in _general_ is much less
         | interesting or helpful than being able to statistically
         | quantify the behavior of the systems we 're building.
         | 
         | The author is a PhD candidate at the Carnegie Mellon School of
         | Business, and I was quite taken with their ability to hop
         | across fields to get a rather simple and important way to
         | systematically and statistically review the systems we're
         | building.
        
           | apstroll wrote:
           | This paper is doing exactly that though, handwaving with a
           | couple of floats. The paper is just a collection of
           | observations about what their implementation of shapley value
           | analysis gives for a few variations of a prompt.
        
         | cauliflower2718 wrote:
         | It looks like it's written in latex to me. Standard formatting
         | varies across departments, and the author is in the business
         | school at CMU.
         | 
         | In some fields, single author papers are more common. Also,
         | outside of ML conference culture, the journal publication
         | process can be pretty slow.
         | 
         | Based on the above (which is separate from an actual evaluation
         | of the paper), there are no immediate red flags.
         | 
         | Source: I am a PhD student and read papers across stats/CS/OR.
        
           | woolion wrote:
           | The Latex feel comes in good part from the respect for
           | typographical standards that is encoded as default behaviour.
           | In this document, so many spacings are just flat-out wrong,
           | first paragraph indents, etc. If it's indeed Latex (it kinda
           | looks like it), someone worked hard to make it look bad.
           | 
           | The weirdest thing is that copy-paste doesn't work; if I copy
           | the "3.1" of the corresponding equation, I get " . "
        
           | ersiees wrote:
           | Another clue: there is no way to download the latex, while
           | you can if someone uploaded the latex on arxiv.
        
             | cauliflower2718 wrote:
             | There's a lazy way to submit to arxiv, which is to submit
             | just the PDF, even if you did it in latex. Sometimes it can
             | be annoying to organize the tex files to submit to arxiv.
             | It's uncommon, but the font and math rendering are the
             | standard latex font.
        
         | mnky9800n wrote:
         | I think that we should not accept peer review as some kind of
         | gold standard anymore for several reasons. These are my
         | opinions based on my experience as a scientist for the last 11
         | years.
         | 
         | - its unpaid work and often you are asked to do it too much and
         | therefore may not give your best effort
         | 
         | - editors want to have high profile papers and minimise review
         | times so glossy journals like nature or science often reject
         | things that require effort on the review
         | 
         | - the peers doing a review are often anything but. I have seen
         | self professed machine learning "experts" not know the
         | difference between regression and classification yet proudly
         | sign their names to their review. I've seen reviewers ask you
         | to write prompts that are mean and cruel to an LLM to see if it
         | would classify test data the same (text data from geologists
         | writing about rocks). As an editor I have had to explain to
         | adult tenured professor that she cannot write in her review
         | that the authors were "stupid" and "should never be allowed to
         | publish again".
        
           | 3abiton wrote:
           | Scientific peer review is another facit of civilization that
           | its current design does not allow it to scale well. More and
           | more people are being involved in the process, but the
           | qualityis forever going down.
        
             | mnky9800n wrote:
             | Yes that's right. It's a scaling problem and there isn't a
             | clear answer. It's easy to complain about it though haha. I
             | think what is happening is science is atomitizing. People
             | are publishing smaller amounts or simply creating ideas
             | from nothing (like that science advances paper on hacker
             | news a couple days ago that created a hydrogen rich crust
             | from thin air).
        
           | chongli wrote:
           | A further issue is peer review quid pro quo corruption. The
           | reviewer loves your paper but requests one small change: cite
           | some of his papers and he'll approve your paper.
           | 
           | I don't know how prevalent this sort of corruption is (I
           | haven't read any statistical investigations) but I have heard
           | of researchers complaining about it. In all likelihood it's
           | extremely prevalent in less reputable journals but for all we
           | know it could be happening at the big ones.
           | 
           | The whole issue of citations functioning like a currency
           | recalls Goodhart's Law [1]:
           | 
           |  _"When a measure becomes a target, it ceases to be a good
           | measure."_
           | 
           | [1] https://en.wikipedia.org/wiki/Goodhart's_law
        
             | mnky9800n wrote:
             | Tbh I used to have an issue with that but these days it
             | really is a small issue in the grand scheme of things. You
             | can say No but also, there are larger systemic problems in
             | science.
        
               | chongli wrote:
               | You're right. It's more of a symptom of the systemic
               | problems than the main problem itself. But it still
               | contributes to my distrust in science.
        
         | chongli wrote:
         | I didn't even read the paper, I just read the abstract. I was
         | really impressed by the idea of using Shapley values to
         | investigate how each token in a prompt affects the output,
         | including order-based effects.
         | 
         | Even if the paper itself is rubbish I think this approach to
         | studying LLMs at least warrants a second look by another team
         | of researchers.
        
         | johndough wrote:
         | Two more heuristics:
         | 
         | 1. The figures are not vectorized (text in figures can not be
         | selected). All it takes is to replace "png" in
         | `plt.savefig("figure.png")` with "pdf", so this is a very easy
         | fix. Yet the author did not bother, which shows that he either
         | did not care or did not know.
         | 
         | 2. The equations lack punctuation.
         | 
         | Of course you can still write insightful papers with low
         | quality figures and unusual punctuation. This is just a
         | heuristic after all.
        
       | xianshou wrote:
       | This doesn't replicate using gpt-4o-mini, which always picks
       | Flight B even when Flight A is made somewhat more attractive.
       | 
       | Source: just ran it on 0-20 newlines with 100 trials apiece,
       | raising temperature and introducing different random seeds to
       | prevent any prompt caching.
        
         | yorwba wrote:
         | The newline thing is the motivating example in the
         | introduction, using Llama 3 8B Instruct with up to 200 newlines
         | before the question. If you want to reproduce this example with
         | another model, you might have to increase the number of
         | newlines all the way to the context limit. (If you ask the API
         | to give you logprobs, at least you won't have to run mutiple
         | trials to get the exact probability.)
         | 
         | But the meat of the paper is the Shapley value estimation
         | algorithm in appendix A4. And in A5 you can see that different
         | models giving different results is to be expected.
        
       | scottiescottie wrote:
       | explainable AI just ain't there yet.
       | 
       | I wonder if the author took a class with Lipton, since he's at
       | CMU. We literally had a lecture about Shapley Values "explaining"
       | AI. It's BS.
        
       ___________________________________________________________________
       (page generated 2024-12-28 23:01 UTC)