The Illusion of Thinking: Strengths and limitations of reasoning models [pdf]
ml-site.cdn-apple.com
The Illusion of Thinking: Strengths and limitations of reasoning models [pdf]
1–10 of 276 posts
Re: The Illusion of Thinking: Strengths and limitations of reasoning models [pdf]
#2Okay Apple, you got my attention. But I'm a strong proponent of "something is better than nothing" philosophy—even if OpenAI/Google/etc. are building reasoning models with the limitations that you describe, they are still a huge progress compared to what we had not long ago. Meanwhile you're not even trying.
It's so easy to criticize the works of others and not deliver anything. Apple—be Sam in Game of Thrones: "I'm tired of reading about the achievements of better men".
Re: The Illusion of Thinking: Strengths and limitations of reasoning models [pdf]
#3[flagged]
Re: The Illusion of Thinking: Strengths and limitations of reasoning models [pdf]
#4[flagged]
Equivalent: "I'm so sick of these [atheist] cretins proclaiming that [god doesn't exist], and when you drill down to first principles, these same people do not have a rigorous definition of [god] in the first place!"
That's nonsense, because the people obligated to furnish a "rigorous definition" are the people who make the positive claim that something specific is happening.
Also, the extraordinary claims are the ones that require extraordinary evidence, not the other way around.
Re: The Illusion of Thinking: Strengths and limitations of reasoning models [pdf]
#5[flagged]
Equivalent: "I'm so sick of these [atheist] cretins proclaiming that [god doesn't exist], and when you drill down to first principles, these same people do not have a rigorous definition of [god] in the first place!" That's nonsense, because the people obligated to furnish a "rigorous definition" are the people who make the positive claim that something specific is happening. Also, the extraordinary claims are the on…
Your god analogy is clumsy in this case. We aren't talking about something fantastical here. Reasoning is not difficult to define. We can go down that road if you'd like. Rather, the problem is, once you do define it, you will quickly find that LLMs are capable of it. And that makes human exceptionalists a bit uncomfortable.
Re: The Illusion of Thinking: Strengths and limitations of reasoning models [pdf]
#6This is easily explained by accepting that there is no such thing as LRMs. LRMs are just LLMs that iterate on its own answers more (or provides itself more context information of a certain type). The reasoning loop on an "LRM" will be equivalent to asking a regular LLM to "refine" its own response, or "consider" additional context of a certain type. There is no such thing as reasoning basically, as it was always a method to "fix" hallucinations or provide more context automatically, nothing else. These big companies baked in one of the hackiest prompt engineering tricks that your typical enthusiast figured out long ago and managed to brand it and profit off it. The craziest part about this was Deepseek was able to cause a multi billion dollar drop and pump of AI stocks with this one trick. Crazy times.
Re: The Illusion of Thinking: Strengths and limitations of reasoning models [pdf]
#7[flagged]
> when you drill down to first principles
It is a trap to consider that first principles perspective is sufficient. Like, if I tell you before 1980: "iterate over f(z) = z^2 + c" there is no way you are going to guess fractals emerge. Same with the rules for Conway's Game of Life - seeing the code you won't guess it makes gliders and guns.
My point is that recursion creates its own inner opacity, it is irreducible, so knowing a recursive system behavior from its iteration code alone is insufficient. It is a becoming not a static thing. That's also why we have the halting problem - recursion.
Reasoning is a recursive process too, you can't understand it from analyzing its parts.
Re: The Illusion of Thinking: Strengths and limitations of reasoning models [pdf]
#8Earlier quoted context omitted.
Equivalent: "I'm so sick of these [atheist] cretins proclaiming that [god doesn't exist], and when you drill down to first principles, these same people do not have a rigorous definition of [god] in the first place!" That's nonsense, because the people obligated to furnish a "rigorous definition" are the people who make the positive claim that something specific is happening. Also, the extraordinary claims are the on…
Your god analogy is clumsy in this case. We aren't talking about something fantastical here. Reasoning is not difficult to define. We can go down that road if you'd like. Rather, the problem is, once you do define it, you will quickly find that LLMs are capable of it. And that makes human exceptionalists a bit uncomfortable.
"It's just statistics!", said the statistical-search survival-selected statistically-driven-learning biological Rube Goldberg contraption, between methane emissions. Imagining that a gradient learning algorithm, optimizing its statistical performance, is the same thing as only learning statistical relationships.
Inconvertibly demonstrating a dramatic failure in its, and many of its kind's, ability to reason.
Re: The Illusion of Thinking: Strengths and limitations of reasoning models [pdf]
#9Okay Apple, you got my attention. But I'm a strong proponent of "something is better than nothing" philosophy—even if OpenAI/Google/etc. are building reasoning models with the limitations that you describe, they are still a huge progress compared to what we had not long ago. Meanwhile you're not even trying. It's so easy to criticize the works of others and not deliver anything. Apple—be Sam in Game of Thrones: "I'm…
I think you're mistaking the work of researchers who work at Apple with the particular investment decisions of Apple over the past few years.
>It's so easy to criticize the works of others and not deliver anything. Apple—be Sam in Game of Thrones: "I'm tired of reading about the achievements of better men".
This is a patently absurd thing to write about a research paper.
Re: The Illusion of Thinking: Strengths and limitations of reasoning models [pdf]
#10Earlier quoted context omitted.
Equivalent: "I'm so sick of these [atheist] cretins proclaiming that [god doesn't exist], and when you drill down to first principles, these same people do not have a rigorous definition of [god] in the first place!" That's nonsense, because the people obligated to furnish a "rigorous definition" are the people who make the positive claim that something specific is happening. Also, the extraordinary claims are the on…
Your god analogy is clumsy in this case. We aren't talking about something fantastical here. Reasoning is not difficult to define. We can go down that road if you'd like. Rather, the problem is, once you do define it, you will quickly find that LLMs are capable of it. And that makes human exceptionalists a bit uncomfortable.
Humans are exceptional though.