Are Emergent Abilities of Large Language Models a Mirage? · HackerPapers discussions

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by hodgehog11

Mon, 28 Sep 2026 13:56:02 GMT

Hallucination rate is a highly nonlinear metric relative to other model success metrics. A similar phenomenon to what is going on here: https://arxiv.org/abs/2304.15004 . This does not mean progress has stalled.

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by nsagent

Mon, 13 Jul 2026 14:45:58 GMT

Then they are naive. I've seen this a lot first hand during my PhD. Interpreting results in a way that reinforces their beliefs and ignores alternate hypotheses. Frequently, those alternative hypotheses are demonstrated to be true. For example, "emergence" as was researched and proclaimed by many labs, including places like OpenAI and Anthropic is due to using too few discrete steps in sampling a continuous phenomenon. See "Are Emergent Abilities of Large Language Models a Mirage?" [1] for example. [1]: https://arxiv.org/abs/2304.15004

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by thesz

Wed, 27 May 2026 14:50:09 GMT

LLMs are Markov Chains [1]. "Emergent abilities" of LLMs can be explained by decrease of perplexity in text prediction [2]. [1] https://arxiv.org/abs/2410.02724 [2] https://arxiv.org/abs/2304.15004

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by thesz

Thu, 02 Apr 2026 19:12:07 GMT

> You recognize that you haven't really needed strong mathematical (or coding) skills to create models for some time. And then there goes something like this [1], where researchers failed to control for p-value: "In this particular setting, emergent abilities claims are possibly infected by a failure to control for multiple comparisons. In BIG-Bench alone, there are ≥220 tasks, ∼40 metrics per task, ∼10 model families, for a total of ∼10^6 task-metric-model family triplets, meaning probability that no task-metric-model family triplet exhibits an emergent ability by random chance might be…

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by yorwba

Thu, 12 Mar 2026 13:14:42 GMT

Yes, I think this is basically an instance of the "emergent abilities mirage." https://arxiv.org/abs/2304.15004 If you measure completion rate on a task where a single mistake can cause a failure, you won't see noticeable improvements on that metric until all potential sources of error are close to being eliminated, and then if they do get eliminated it causes a sudden large jump in performance. That's fine if you just want to know whether the current state is good enough on your task of choice, but if you also want to predict future performance, you need to break it down into smaller…

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

Are Emergent Abilities of Large Language Models a Mirage? — HN submission

Fri, 27 Feb 2026 01:00:02 GMT

Are Emergent Abilities of Large Language Models a Mirage? (2023)

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by charcircuit

Wed, 07 Jan 2026 08:43:24 GMT

>the acceptance that there are emergent properties which appear out of nowhere is another way of saying our scaling laws don’t actually equip us to know what is coming. Is this actually accepted? Ever since [0], I thought people recognized that they don't appear out of nowhere. [0] https://arxiv.org/pdf/2304.15004

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by exasperaited

Thu, 14 Aug 2025 22:06:13 GMT

Which we probably should not, at least not the "sudden" emergence that those researchers claimed to see. https://arxiv.org/abs/2304.15004 Good article about why here; this helped me understand a lot: https://www.wired.com/story/how-quickly-do-large-language-mo...

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by cratermoon

Thu, 08 May 2025 20:32:18 GMT

Alternate view: Are Emergent Abilities of Large Language Models a Mirage? https://arxiv.org/abs/2304.15004 "Here, we present an alternative explanation for emergent abilities: that for a particular task and model family, when analyzing fixed model outputs, emergent abilities appear due to the researcher's choice of metric rather than due to fundamental changes in model behavior with scale. Specifically, nonlinear or discontinuous metrics produce apparent emergent abilities, whereas linear or continuous metrics produce smooth, continuous predictable changes in model performance."

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by godelski

Sat, 08 Feb 2025 01:19:33 GMT

I don't think this makes sense and I'm not quite sure why you went to ML, but that's okay. I am a machine learning researcher, but also frustrated with the state of machine learning, in part because, well... you can probably see how "proof by empirical evidence" is dialed up to 11. Sorry, long answer incoming. It is far from complete too but I think it will help build strong intuition around your questions. Will knowledge transfer? That entirely depends on the new problem. It also entirely depends on how related the problem is. But also, what information was used to solve the pre-transfer…

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by abernard1

Fri, 13 Sep 2024 05:43:40 GMT

I'll give an argument against this with the caveat it applies only if these are pure LLMs without heuristics or helper models (I do not believe that to be the case with o1). The problem with auditing is not only are the outputs incorrect, but the "inputs" of the chained steps have no fundamental logical connection to the outputs. A statistical connection yes, but not a causal one. For the trail to be auditable, processing would have to be taking place at the symbolic level of what the tokens represent in the steps. But this is not what happens. The transformer(s) (because these are now…

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by humansareok1

Thu, 04 Apr 2024 14:50:46 GMT

>> To study how the parallel structure of transformers might limit their capabilities, the pair considered the case where transformers didn’t feed their output back into their input — instead, their first output would have to be the final answer. They proved that the transformers in this theoretical framework couldn’t solve any computational problems that lie outside a specific complexity class. And many math problems, including relatively simple ones like solving linear equations, are thought to lie outside this class. This seems like somewhat of an artificial constraint? E.g. we can…

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by nicklecompte

Wed, 03 Apr 2024 15:16:39 GMT

There was an article on HN last week discussing theoretical proof that scaling has a hard limit: https://www.quantamagazine.org/how-chain-of-thought-reasonin... Specifically: > To study how the parallel structure of transformers might limit their capabilities, the pair considered the case where transformers didn’t feed their output back into their input — instead, their first output would have to be the final answer. They proved that the transformers in this theoretical framework couldn’t solve any computational problems that lie outside a specific complexity class. And many math problems…

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by n2d4

Mon, 25 Mar 2024 05:31:34 GMT

The original emergence paper has multiple "emergent" metrics. [1] Though, the paper this article is about [2] mentions that among the "emergent" tasks they tested, 92% were measuring accuracy on either exact-string-match or multiple choice, so probably just bad metrics. [1] https://arxiv.org/pdf/2206.07682.pdf [2] https://arxiv.org/pdf/2304.15004.pdf

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by n2d4

Mon, 25 Mar 2024 05:08:57 GMT

It depends on how you do arithmetic. If you're a human and you do column addition, 12345+35791=58136 is just as big of a mistake as 48146 (the actual result is 48136). It's just one mistaken column in both. Binary half-adders work the same way. We don't really know how LLMs do arithmetic. Maybe token edit distance would be interesting, but either way it doesn't really change the claim of the paper. Unrelated: The link is incorrect, the one you're referring to is here: https://arxiv.org/pdf/2304.15004.pdf

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by esafak

Mon, 25 Mar 2024 01:45:08 GMT

The paper: Are Emergent Abilities of Large Language Models a Mirage? https://arxiv.org/abs/2304.15004

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by nyrikki

Tue, 19 Mar 2024 18:28:55 GMT

>Yes, we don't really understand where emergent capabilities are coming from, at least not to extent of being able to predict them ahead of time ("if we feed it this amount of data, of this type, it'll learn to do X"). New emergent capabilities arise, from time to time, as models are scaled up, but no one can predict exactly what their next-gen model is going to be capable of. While finite precision, finite width transformers aren't TC, I don't see why the same property of the game of life, where one cannot predict the end state from the starting state wouldn't hold. As we know transformers…

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by nicklecompte

Fri, 01 Mar 2024 16:37:00 GMT

Perhaps you should update your priors about "emergent behavior" in GPT3+: https://arxiv.org/abs/2304.15004

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by nsagent

Wed, 27 Dec 2023 13:50:11 GMT

I mentioned this to another commenter as well: You might want to reconsider your stance on emergent abilities in LLMs considering the NeurIPS 2023 best paper winner is titled: "Are Emergent Abilities of Large Language Models a Mirage?" https://arxiv.org/abs/2304.15004 https://blog.neurips.cc/2023/12/11/announcing-the-neurips-20...

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by nsagent

Wed, 27 Dec 2023 13:47:25 GMT

You might want to reconsider your stance on emergent abilities in LLMs considering the NeurIPS 2023 best paper winner is titled: "Are Emergent Abilities of Large Language Models a Mirage?" https://arxiv.org/abs/2304.15004 https://blog.neurips.cc/2023/12/11/announcing-the-neurips-20...

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by nicklecompte

Tue, 26 Dec 2023 15:48:11 GMT

AI is not even close to having true logical reasoning, that's probably decades away. The issue is that cognitive scientists are clueless. Scientists have a good model for associative reasoning, which is the basis of modern neural networks, but we don't have a clue how abstract reasoning actually works. All birds and mammals have advanced abstract reasoning and are far more intelligent than GPT-4: - birds and mammals are inherently able to count in almost any context because they understand what numbers actually mean; GPT-4 can only be trained to count in certain contexts. GPT-4 would be like…

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by sgt101

Fri, 15 Dec 2023 16:07:49 GMT

No emergence [2304.15004] Are Emergent Abilities of Large Language Models a Mirage? - arXiv https://arxiv.org/abs/2304.15004 Can't plan https://openreview.net/forum?id=X6dEqXIsEW No compositionality https://openreview.net/forum?id=Fkckkr3ya8 Apart from that it's great

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by pmontra

Tue, 12 Dec 2023 13:01:01 GMT

The paper is at https://arxiv.org/abs/2304.15004 The submitted link is to a prizegiving conference where the authors will be (probably? I'm not sure) presented with an award.

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by alephxyz

Tue, 31 Oct 2023 20:44:39 GMT

>While performance on benchmarks typically improves smoothly, sometimes specific capabilities emerge without warning (Wei et al., 2022a). The conclusions of that paper aren't very convincing (see https://arxiv.org/abs/2304.15004 ).

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by photonthug

Sat, 23 Sep 2023 00:20:05 GMT

Idk, one obvious reason could be that emergence is a mirage ( https://arxiv.org/abs/2304.15004 ). Given two concepts of roughly equivalent complexity and representation in training, where one is successfully generalized and another is not, that would seem to support (but not prove) the mirage hypothesis. I take your point about representational differences though. Thinking more about this, the OP is basically talking about whether models understand that equality commutes, and I suppose even an LLM that can add may not have worked out that addition commutes.

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by ianbutler

Sun, 18 Jun 2023 17:24:55 GMT

The paper was fairly recent, https://arxiv.org/abs/2304.15004. It was more thoroughly confirming what was generally agreed upon while debunking other reasons. The key insight from the abstract, "Specifically, nonlinear or discontinuous metrics produce apparent emergent abilities, whereas linear or continuous metrics produce smooth, continuous predictable changes in model performance." Yup that is another recent and interesting development for sure!

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by srslack

Thu, 18 May 2023 21:05:41 GMT

Wow, this is refreshing. To your point - the underpinning technology (regression based function approximation) holds more importance than general and adaptable intelligence. It holds more importance than something that is going to, or is capable of, "escaping the box and killing us all." "Emergent behavior", when it's not just a mirage or poor word choice of wishful researchers (that it does things previous models did not do, iirc, very poor word choice - https://arxiv.org/abs/2304.15004) and if it even could exist, is only a side effect of the regression based function approximation to…

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by revelio

Tue, 16 May 2023 19:26:59 GMT

"Are Emergent Abilities of Large Language Models a Mirage?" https://arxiv.org/pdf/2304.15004.pdf our alternative suggests that existing claims of emergent abilities are creations of the researcher’s analyses, not fundamental changes in model behavior on specific tasks with scale.

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by popinman322

Mon, 08 May 2023 23:44:19 GMT

There was a discussion about the paper[0] this was based on 7 days ago: https://news.ycombinator.com/item?id=35768824 [0] https://arxiv.org/abs/2304.15004

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

Are Emergent Abilities of Large Language Models a Mirage? — mentioned by eternalban

Wed, 03 May 2023 12:22:53 GMT

The Stanford paper's title ends with a question mark. Are Emergent Abilities of Large Language Models a Mirage?, (2023) https://arxiv.org/abs/2304.15004

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

Are Emergent Abilities of Large Language Models a Mirage? — HN submission

Mon, 01 May 2023 03:32:48 GMT

Are emergent abilities of large language models a mirage?

154 points; 130 discussion replies on HN. Paper and source history. A submission is not an endorsement.