LMArena is a cancer on AI

199 points | by jumploops a day ago

85 comments

  • atleastoptimal 5 hours ago

    The general conceit of this article, which is something that many frontier labs seem to be beginning to realize, is that the average human is no longer smart enough to provide sufficient signal to improve AI models.

      gpm 2 hours ago

      No, it's that the average unpaid human doesn't care to read closely enough to provide signal to improve AI models. Not that they couldn't if they put in even the slightest amount of effort.

        kazinator 2 hours ago

        Firstly, paying is not at all the correct incentive for the desired outcome. When the incentive is payment, people will optimize for maximum payout not for the quality goals of the system.

        Secondly, it doesn't fix stupidity. A participant who earnestly takes the quality goals of the system to heart instead of focusing on maximizing their take (thus, obviously stupid) will still make bad classifications due to that reason.

          tbrownaw an hour ago

          > Firstly, paying is not at all the correct incentive for the desired outcome. When the incentive is payment, people will optimize for maximum payout not for the quality goals of the system.

          1. I would expect any paid arrangement to include a quality-control mechanism. With the possible exception of if it was designed from scratch by complete ignoramuses.

          2. Do you have a proposal for a better incentive?

        ehnto 2 hours ago

        Why would an unpaid human want to do that?

          0manrho 10 minutes ago

          Therein lies the problem.

          alterom 2 hours ago

          Exactly — they wouldn't.

      Y_Y 5 hours ago

      But when you're a moron how can you distinguish?

      I'm being (mostly) serious, suppose you're a stuffed ahort trying to boost your valuation, how can you work out who's smart enough to train your LLM? (Never mind how to get them to work for you!)

        aspenmartin 5 hours ago

        I do a lot of human evaluations. Lots of Bayesian / statistical models that can infer rater quality without ground truth labels. The other thing about preference data you have to worry about (which this article gets at) is: preferences of _who_? Human raters are a significantly biased population of people, different ages, genders, religions, cultures, etc all inform preferences. Lots of work being done to leverage and model this.

        Then for LMArena there is the host of other biases / construct validity: people are easily fooled, even PhD experts; in many cases it’s easier for a model to learn how to persuade than actually learn the right answers.

        But a lot of dismissive comments as if frontier labs don’t know this, they have some of the best talent in the world. They aren’t perfect but they in a large sene know what they’re doing and what the tradeoffs of various approaches are.

        Human annotations are an absolute nightmare for quality which is why coding agents are so nice: they’re verifiable and so you can train them in a way closer to e.g. alphago without the ceiling of human performance

          fc417fc802 5 hours ago

          > in many cases it’s easier for a model to learn how to persuade than actually learn the right answers

          So we should expect the models to eventually tend toward the same behaviors that politicians exhibit?

            c0balt 4 hours ago

            Maybe a happy to deceive marketing/sales role would be more accurate.

          RA_Fisher 4 hours ago

          100% (am a Bayesian statistician).

          Isn’t it fascinating how it comes down to quality of judgement (and the descriptions thereof)?

          We need an LMArena rated by experts.

            Lerc 16 minutes ago

            As a statistician, do you you think you could, given access to the data, identify the subset of LMArena users that are experts?

          zqy123007 2 hours ago

          they always know, they just have non-AGI incentive and asymetric upside to play along...

        wongarsu 5 hours ago

        Sure, on the surface judging the judge is just as hard as being the judge

        But at least the two examples of judging AI provided in the article can be solved by any moron by expending enough effort. Any moron can tell you what Dorothy says to Toto when entering Oz by just watching the first thirty minutes of the movie. And while validating answer B in the pan question takes some ninth-grade math (or a short trip to wikipedia), figuring out that a nine inch diameter circle is in fact not the same area as a 9x13 inch square is not rocket science. And with a bit of craft paper you could evaluate both answers even without math knowledge

        So the short answer is: with effort. You spend lots of effort on finding a good evaluator, so the evaluator can judge the LLM for you. Or take "average humans" and force them to spend more effort on evaluating each answer

          michaelmrose 2 hours ago

          Maybe you need to have people rate others ratings to remove at least the worst idiots.

        atleastoptimal 5 hours ago

        that’s why Mercor is worth 2billion

      ryandrake 4 hours ago

      Popularity has never been a meaningful signal of quality, no matter how many tech companies try to make it so, with their star ratings, up/down voting, and crowdsourcing schemes.

        PaulHoule 3 hours ago

        Different strokes for different folks: I mean who is to say if Bleach or Backstabbed in a Backwater Dungeon: My Trusted Companions Tried to Kill Me, but Thanks to the Gift of an Unlimited Gacha I Got LVL 9999 Friends and Am Out for Revenge on My Former Party Members and the World is better?

      kazinator 2 hours ago

      It is glaringly obvious that the average human is not smart enough to the level hat their decision making should be replicated and adopted at scale.

      People hold falsehoods to be true, and cannot calculate a 10% tip.

      Yizahi 5 hours ago

      Yep, it's like getting a commoner from the street evaluate a literature PhD in their native language. Sure, both know the language, but the depth difference of a specialist vs a generalist is too large. And neither we can't use AI to automatically evaluate this literature genius because real AI doesn't exist (yet), hence the programs can't understand the contents of text they output or input. Whoops. :)

      michaelmrose 2 hours ago

      The average human is a moron you wouldn't trust to watch your hamster. If you watched them outside of the narrow range of tasks they have been trained to perform by rote you would probably conclude they should qualify for benefits by virtue of mental disability.

      We give them WAY too much credit by watching mostly the things they have been trained specifically to do and pretending this indicates a general mental competence that just doesn't exist.

      echelon an hour ago

      If these frontier models were open source, the market of downstream consumers would figure out how to optimize them.

      By being closed, they'll never be optimal.

      cyanydeez 5 hours ago

      They need to spend money on actual experts to curate their data to improve.

      Instead, finance bros are convinced by the argument that number goes up.

        8f2ab37a-ed6c 5 hours ago

        Is that not exactly what https://www.mercor.com/ does?

        aspenmartin 5 hours ago

        Wait you know that frontier labs do actually do this right?

        Terr_ 5 hours ago

        Sometimes it feels like:

            def is_it_true(question): 
                return profit_if_true(question) > profit_if_false(question)
        
        AI will make it cheaper, faster, better, no problem. You can eat the cake now and save it for later.
  • fuddle 4 hours ago

    > It's past time for LMArena people to sit down and have some thorough reflection on whether it is still worth running at all

    They've raised about $250 million, so I don't see that happening anytime soon.

      londons_explore 4 hours ago

      I kinda assumed they wouldn't need any money because AI companies give them free credits to evaluate the models, and users ask questions and rate for free because they get to use decent AI models at no cost...

      Beyond that there is coding up a web page, which as we all know can be vibe coded in a few hours...

      What else is there to spend money on?

  • observationist 6 hours ago

    There's something deeply ironic about this being written by AI. Baitception, even.

      denismi 4 hours ago

      "The Brutal Choice"

      Is there an established name for this LLMism?

      I don't need a "Reality Check" or a "Hard Truth". The thought can be concluded without this performative honesty nonsense or the emotive hyperbole.

      This probably grates me more than any other.

      dust42 5 hours ago

      Oh my goodness yes, I almost missed it that the text is (mostly?) AI written. That said I agree that LMArena elo scores are pushing models in the wrong direction. They move more towards McDonald's than quality food.

      duncancarroll 3 hours ago

      This was my first thought as well

      aratahikaru5 2 hours ago

      How can you tell? (honest question, I really can't)

      The article makes strong points, includes real data and quotes, shows proof of work (sampling 100 Q&A), so does that even matter at this point? This doesn't feel like "slop" to me at all.

        ryan_n 2 hours ago

        Yea I also didn't think this was written by ai, it sounded human enough to me. It's kind of a bummer that there's all these patterns that LLM's follow in their output that cause people to have a knee jerk reaction and instantly call it ai slop. I know there is a ton of ai garbage out there these days, but I really couldn't tell with this article.

        joe_the_user an hour ago

        The text definitely the "jump from dramatic crescendo to dramatic crescendo" quality of certain LLM texts. If you read closely, it also has adjective choice that's more for dramatic than appropriate to the circumstances involves (a quality of LLM texts it also helpfully explains).

        I don't know if this proves it's an LLM text or whether that style is simply spilling out everywhere.

  • aucisson_masque 5 hours ago

    > They're not reading carefully. They're not fact-checking, or even trying.

    It’s not how I do, and I suppose how many people do. I specifically ask questions related to niche subjects that I know perfectly well and that is very easy for me to spot mistakes.

    The first time I used it, that’s what came naturally to my mind. I believe it’s the same for others.

      Sharlin 2 hours ago

      Unfortunately I don't think there's any reason to assume that you're a representative sample of LMArena users.

      p-e-w 4 hours ago

      Yeah, that quote just reads like the typical “everyone is an idiot except me” attitude that pervades the tech world.

      Of course people visiting a website specifically designed for evaluating LLMs do try all kinds of specific things to specifically test for weaknesses. There may be users who just click on the response with more emojis, but I strongly doubt they are the majority on that particular site.

  • stared 5 hours ago

    When they released GPT-4.5, it was miles ahead of others when it comes to its linguistic skills and insight. Yet, it was never at top of the arena - it felt that not everone was able to appreciate the edge.

      johnsmith1840 4 hours ago

      4.5 was easily the best conversationalist I've seen. Not as powerful as modern ones but something about HOW it talked felt inherently smart.

      I miss that one, is 5 any better? I switched to claude before it launched.

        Vecr 3 hours ago

        > something about HOW it talked felt inherently smart

        The thing was huge. They were training the thing to be GPT5, before they figured out their userbase to too large to be served something that big.

        kingstnap 2 hours ago

        No replacement for displacement, except applied to LLMs and raw parameter count.

  • aipatselarom 3 hours ago

    >Would you trust a medical system measured by: which doctor would the average Internet user vote for?

    Yes, the system desperately needs this. Many doctors malpractice for DECADES.

    I would absolutely seek to, damn, even pay good money to, be able to talk with a doctor's previous patients, particularly if they're going to perform a life-changing procedure on me.

      stonogo 3 hours ago

      Doctors would also pay good money for votes, so I'm not sure that would fix anything.

        michaelmrose 2 hours ago

        Raw score is often quite frankly crap. It's often still easy to surface the negative reviews and since people don't at least at present fake those you can find out what they didn't like about a product. If a given products critics are only those whining about something irrelevant, not meaningful to your use case, or acceptable to you and it overall appears to meet spec you are often golden.

  • dk8996 6 hours ago

    Seems like they just raised 150m at 1.7B valuation. Crazy.

      minimaxir 5 hours ago
      koakuma-chan 6 hours ago

      Who? LMArena? That's actually crazy.

        echelon 5 hours ago

        Are they selling:

        A. model improvement tests, suites, and benchmarks

        B. data on competitors' evals

        C. test answer keys

        D. alpha to VC firms

        E. all of the above

        ???

          koakuma-chan 5 hours ago

          Apparently they are selling model evaluations, powered by their volunteer users.

            ares623 18 minutes ago

            They're selling "I'm an AI investor" stickers to show off at the next family reunion

            Y_Y 5 hours ago

            I'm taking the Red Cross public next. With the price of healthcare these days my earnings projections are uber-extreme.

  • tbrownaw an hour ago

    From https://lmarena.ai/how-it-works:

    > In battle mode, you'll be served 2 anonymous models. Dig into the responses and decide which answer best fits your needs.

    It's not a given that someone's needs are "factual accuracy". Maybe they're after entertainment, or winning an argument.

  • sharkjacobs 6 hours ago

    Any metric that can be targeted can be gamed

      kelseyfrog 5 hours ago

      Then target it with metrics worth solving[1].

      1. Ex https://mppbench.com/

        falcor84 5 hours ago

        But that seems to be measuring "superintelligence" rather than just AI, no?

      positron26 4 hours ago

      If the metric is a latent variable summarizing subjective judgements, yes.

  • zemo 4 hours ago

    this argument is also broadly true about the quality and correctness of posts on any vote-based discussion board

    > Why is LMArena so easy to game? The answer is structural. > The system is fully open to the Internet. LMArena is built on unpaid labor from uncontrolled volunteers.

    also all user's votes count equally, bu not all users have equal knowledge.

      coderenegade 3 hours ago

      As long as users are better than 50% accurate, it shouldn't matter if they're experts or not. That being said, it's difficult to measure user accuracy in this case without running into circular reasoning.

  • mirekrusin 5 hours ago

    True and what you can realize/read between the lines is something deeper.

    LLMs are fallible. Humans are fallible. LLMs improve (and improve fast). Humans do not (overall, ie. "group of N experts in X", "N random internet people").

    All those "turing tests" will start flipping.

    Today it's "N random internet humans" score too low on those benchmarks, tomorrow it'll be "group of N expert humans in X" score too low.

  • kazinator 2 hours ago

    The average person is dumber than an LLM in terms of having a grasp on the facts, and basic arithmetic.

    A voting system open to the public is completely screwed even if somehow its incentives are optimized toward strongly encouraging ideal behavior.

  • boredemployee 2 hours ago

    > Being verbose. Longer responses look more authoritative!

    I know we can solve this in ordinary tasks just using prompt but that's really annoying. Sometimes I just want a yes or no answer and then I get a phd thesis in the matter.

  • jpollock 3 hours ago

    Couldn't "The Wisdom of Crowds" help with this?

    Maybe if they started ranking the answers on a 1-10 range, allowing people to specify graduations of correctness/wrongness, then the crowd would work?

    https://en.wikipedia.org/wiki/The_Wisdom_of_Crowds

  • kahnclusions an hour ago

    AI is a cancer on humanity

  • usef- 5 hours ago

    When the Meta cheating scandal happened I was surprised how little of the attention was on this.

    Meta "cheated" on lmarena not by using a smarter model but by using one that was more verbose and friendly with excessive emojis.

  • BrenBarn 3 hours ago

    Since AI is itself a cancer, maybe this is good? The cancer of my cancer is my chemo.

  • derac 3 hours ago

    Is there any reason to believe LMArena isn't botted by the people releasing these models?

  • fzysingularity 4 hours ago

    > It's like going to the grocery store and buying tabloids, pretending they're scientific journals.

    This is pure gold. I've always found this approach of evals on a moving-target via consensus broken.

  • thorum 5 hours ago

    Aside from Meta is there any reason to think the big AI labs are still using LMArena data for training? The weaknesses are well understood and with the shift to RL there are so many better ways to design a reward function.

      nl 5 hours ago

      I don't think anyone has ever used it as training. But yes labs still do seem to target it as goal (which is a different thing).

      dk8996 5 hours ago

      Such as?

  • big_toast 4 hours ago

    Is there a reason wrong data isn't considered more broadly in its context as still valuable?

    Shouldn't the model effectively 1. learn to complete the incorrect thing and 2. learn the context that it's correct and incorrect? In this case the context being lazy LMArena users. And presumably, in the future, poorly filtered training data.

    We seem to be able to read incorrect things and not be corrupted (well, theoretically). It's not ideal, but it seems an important component to intellectual resilience.

    It seems like the model knowing the data is LMArena, or some type of un-trusted, would be sufficient to shift the prior to a reasonable place.

  • keketi 6 hours ago

    We need a service that ranks AI model ranking services. Maybe powered by AI instead of humans?

      echelon 5 hours ago

      Just look at Open(ugh)Router. That's a good, though not fully accurate, view of where dollars are going.

      It'd be nice if it were actually open and we could inspect all the statistics.

  • a-dub 6 hours ago

    maybe it would work if they could encourage end users to be rigorous? (ie, detect if they have the capability to rate well and then reward them when they do by comparing them against other highly rated raters of the same phenotype)

  • bigdict 2 hours ago

    > What actually happens: random Internet users spend two seconds skimming, then click their favorite.

    > They're not reading carefully. They're not fact-checking, or even trying.

    Uhhh, how was that established?

  • g947o 6 hours ago

    > Voilà: bold text, emojis, and plenty of sycophancy – every trick in the LMArena playbook! – to avoid answering the question it was asked.

    This is hard to swallow.

    I don't believe a single word this article says. Apparently the "real author" (the human being who wrote the original prompt to generate this article) only intend to use this article to generate clicks and engagement but don't care at all about what's in there.

  • alfalfasprout 4 hours ago

    and AI is a cancer on humanity... this article is clearly LLM written too.