Teaching Algorithmic Reasoning via In-context Learning · HackerPapers discussions

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Fri, 15 Dec 2023 02:30:05 GMT

>Every human is not even capable of reading, or speaking, or thinking at all. But those humans who can't are not the benchmark of general human intelligence. It's simple. If you're going to disqualify a potential general intelligence because of a particular task solving ability, it better be something every general intelligence can do. Now let's forget every comparison with the blind, disabled and so on. Let's compare with humans who according to you are "capable of reading, or speaking, or thinking at all ", what's the magic number? Do you really think every human capable of doing all three…

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Mon, 09 Oct 2023 21:00:57 GMT

> Numbers (LLMs still can't do math and the ones that "can" are very brute forced and don't generalize) These seem to generalize. https://arxiv.org/abs/2211.09066 https://arxiv.org/abs/2308.00304 > Commutative properties (The (idk why anyone was surprised) "reversal curse": https://arxiv.org/abs/2309.12288) This was a retrieval from training issue not an inference one. and make no mistake, people will fail these sort of answers too (perhaps not to the same extent). Very often someone will learn a fact in a certain why and fail to recall that when asked in a reverse or roundabout way.…

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Fri, 18 Aug 2023 15:14:49 GMT

LLMs are a different beast in the ML world. Finetuned Palm for Medicine and Finetuned Minerva for Math all perform a good deal worse than GPT-4. A fine-tuned smaller model is by no means guaranteed to beat a larger more general one (though of course you may get acceptable performance). And then the necessity of fine-tuning itself is called into question plenty with LLMs. https://huggingface.co/papers/2308.00304 https://huggingface.co/papers/2308.07921 https://arxiv.org/abs/2211.09066

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Tue, 11 Jul 2023 16:29:54 GMT

>No, but I believe this is a different question. arithmetic but in a different base is one of the counterfactual examples in the paper. That's why i mentioned that. and yes it can answer them with worse performance. You can juice arithmetic performance as is with algorithmic instructions. https://arxiv.org/abs/2211.09066. I see no reason the same for other bases wouldn't work. Even if you gave a human substantial time for it (say a week of study), i believe he/she almost certainly reach the same accuracy unless he had access to specific instructions for working in base 8 he/she could call…

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Wed, 05 Jul 2023 22:41:37 GMT

It's alright with algorithmic prompts - https://arxiv.org/abs/2211.09066 also it knows when to use a calculator if it has access to one so it's not a big deal

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Tue, 13 Jun 2023 04:10:42 GMT

Post moving. "Reasonable" is just an arbitrary line. Especially since most if not all would make some mistake somewhere along the line. You can greatly increase GPT's arithmetic capabilities tackling it like a problem to solve "on paper" in context. And this was done on 3.5 not 4. https://arxiv.org/abs/2211.09066

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by chaxor

Thu, 25 May 2023 14:51:42 GMT

They are GNNs with attention as the message passing function and additional concatenated positional embeddings. As for reasoning, these are not quite 'problems well-trained for', in the sense that they're not in the training data. But they are likely problems that have some abstract algorithmic similarity, which is the point. I'm not quite sure what you mean that they cannot do causal graph analysis, since that was one of many different tasks provided in the various different types of reasoning studies in the paper I mentioned. In fact it may have been the best performing task. Perhaps try…

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Mon, 22 May 2023 23:25:33 GMT

I don't think you took more than a passing glance, if any at those papers. What you describe is impossible with these 3. https://arxiv.org/abs/2212.09196 - new evaluation set introduced with the paper. modelled after tests that previously only had visual equivalents. contamination literally impossible https://arxiv.org/abs/2204.02329 - effect of explanations on questions introduced with the paper. dataset concerns make no sense. https://arxiv.org/abs/2211.09066 - new prompting method introduced to improve algorithmic calculations. dataset concerns make no sense. The Casual paper is the only…

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Mon, 22 May 2023 22:23:07 GMT

>It's not better at reasoning. It's barely even capable of it You are wrong. and there's many papers to show otherwise. Algorithmic, Casual, Inference, Analogical LLMs reason just fine https://arxiv.org/abs/2212.09196 https://arxiv.org/abs/2305.00050 https://arxiv.org/abs/2204.02329 https://arxiv.org/abs/2211.09066

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Sun, 21 May 2023 02:46:45 GMT

No world model ? A world model is so obvious papers like these are more confirmation than surprise https://arxiv.org/abs/2305.11169 https://arxiv.org/abs/2210.13382 There a certain sentiment that AGI however you wish to define it won't infact be a "We'll know it when we see it" situation but rather a "AGI will arrive long before consensus reaches its AGI". LLMs have made me believe this will 100% be the case, either way. It's one thing to argue over things we can't evaluate even now but man the 100th "They can't reason!" every week is pretty funny when you can basically take your pick of…

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Tue, 09 May 2023 01:03:35 GMT

LLMs are not very bad at reasoning. People say this but struggle to demonstrate how. There are many types of reasoning. For some like abstract analogical reasoning, LLMs are top % performers. https://arxiv.org/abs/2212.09196 There's many papers at this point demonstrating pretty strong reasoning. From Casual Reasoning, and explanatory reasoning to algorithmic reasoning. https://arxiv.org/abs/2305.00050 https://arxiv.org/abs/2204.02329 https://arxiv.org/abs/2211.09066 The only kind of reasoning that's somewhat lacking is logical reasoning. And even then it's not awful, just lagging behind…

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by PartiallyTyped

Sat, 29 Apr 2023 16:37:19 GMT

> But it fails at simple arithmetic. Does it though? When allowing LLMs to use their outputs as a form of state they can very much succeed up to 14 digits with > 99.9% accuracy, and it goes up to 18 without deteriorating significantly [1]. That really isn't a good argument because you are asking it to do one-shot something that 99.999% of humans can't. https://arxiv.org/abs/2211.09066

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by shawntan

Tue, 18 Apr 2023 12:30:04 GMT

Yeah the issue tends to be as the length increases the performance drops off. Your approach reminds me of Hattie's work: https://arxiv.org/pdf/2211.09066.pdf I do wonder if the observable drop-off in performance on these formal languages / arithmetic tasks is something more fundamental, and related to the attention mechanism; namely that it can't attend further back as the input gets longer (possibly due to the way attention scores are computed).

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Wed, 12 Apr 2023 21:09:53 GMT

Not too shocking for me after this paper. https://arxiv.org/abs/2211.09066 You can teach GPT-3 arithmetic - https://imgur.com/a/w3DAYOi Basically 100% accuracy up to about 13 digit addition and >90 after that. What else can you teach GPT without changing weights ?

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Wed, 12 Apr 2023 14:39:30 GMT

We are already there. 100% accuracy on up to 13 digit addition can be taught to 3.5 as is. https://arxiv.org/abs/2211.09066 And 4 has little need for such out the box

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Mon, 10 Apr 2023 19:14:24 GMT

In context learning is hands down the biggest breakthrough of LLMs. The flexibility the model displays without updating weights is genuinely mind blowing, bordering on absurd especially if you've trained other kinds of models before. See here - https://imgur.com/a/w3DAYOi from the paper - https://arxiv.org/abs/2211.09066 GPT 3.5's (4 is much much better) addition accuracy tanks after 2 digits. However, by approaching arithmetic as an algorithm to be performed and taught similarly to how it's done with people, you can supercharge accuracy to basically 100% for up to 13 digit addition and >90%…

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Wed, 29 Mar 2023 05:42:34 GMT

The idea that updating weights is particularly necessary is questionable. Memory augmented llms are computationally universal. https://arxiv.org/abs/2301.04589 and make no mistake, in context learning can be used in a meaningful way to resolve deficiencies. gpt-3 getting 98.5% accuracy (on even very large numbers) on addition arithmetic by describing the algorithm of addition to be performed on 2 numbers. https://arxiv.org/abs/2211.09066 In context learning is just overpowered. It's essentially implicit fine-tuning https://arxiv.org/abs/2212.10559v2 as for asking questions, llms can do that…

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Mon, 27 Mar 2023 01:15:35 GMT

I didn't say it's not there. or that you couldn't communicate the ideas of math/arithmetic with language. You can get GPT-3 (yes 3) to have 98.5% accuracy on addition arithmetic (even very large numbers) by..simply describing the algorithm of addition to performed on 2 numbers. https://arxiv.org/abs/2211.09066 The basic idea i'm communicating is that not everything that can be extracted from self supervisory token prediction can be extracted with the same ease. and it is very easy to see why math is one the higher difficulty things. It is extremely easy comparatively to infere "happy" from…

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Mon, 27 Mar 2023 00:03:33 GMT

Oh boy, if you want to fling papers/research, i can do that just fine. https://arxiv.org/abs/2212.09196v1 https://arxiv.org/abs/2211.09066 https://thegradient.pub/othello/ Moreover, "have trouble reasoning" and "can't reason at all" are two very different statements. Not that i particularly agree with either but you should keep that in mind.

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Sun, 26 Mar 2023 16:09:33 GMT

The point I'm making is that chatGPT doesn't know that because the model it has of arithmetic is faulty. It's really as simple as that. The primes thing was just an illustrative example. You want it to do large scale arithmetic with high accuracy? Describe arithmetic as an algorithm to be performed on two numbers. https://arxiv.org/abs/2211.09066 It's not about chatGPT not understanding anything. It's about chatGPT not understanding math very well. That bleeds into understanding of related concepts as well. LLMs build all these models of the world from the text they train on. Well not all if…

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Sun, 26 Mar 2023 14:50:41 GMT

You can increase the accuracy of arithmetic by simply describing arithmetic (addition, subtraction etc) as an algorithm to be performed on the numbers. 98.5% accuracy on addition arithmetic this way with GPT-3 https://arxiv.org/abs/2211.09066

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by dragonwriter

Fri, 17 Mar 2023 06:29:11 GMT

> Yet if you define the algorithmic steps to perform addition on 2 numbers, accuracy on addition arithmetic shoots up to 98% even on very large numbers. https://arxiv.org/abs/2211.09066 Think about what that means. That means that even with the giant model, you need to stuff even the most basic knowledge for dealing with problems of that class into the prompt space to get it to work, cutting into conversation depth and per-response size? The advantage of GPT-4’s big window and the opportunity it provides for things like retrieval and deep iterative context shrinks if I’ve got to stuff a…

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Fri, 17 Mar 2023 03:22:22 GMT

GPT-3 is horrible at arithmetic. Yet if you define the algorithmic steps to perform addition on 2 numbers, accuracy on addition arithmetic shoots up to 98% even on very large numbers. https://arxiv.org/abs/2211.09066 Think about what that means. "Mutating the system" is not a crucial requirement at all. In context learning is extremely over-powered.

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by famouswaffles

Wed, 08 Mar 2023 22:33:47 GMT

If you simply describe addition as an algorithm and ask language models to perform said algorithm on 2 numbers, addition arithmetic accuracy even on very large numbers shoots up to 98.5% https://arxiv.org/abs/2211.09066 LLMs don't struggle with arithmetic because "they can't understand anything!". They struggle with arithmetic because it's really not something that's all that well encoded in language. Self supervision means you aren't necessarily always picking up the most accurate world models. There are models of arithmetic in there. They're just wrong or faulty.

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

Teaching Algorithmic Reasoning via In-context Learning — mentioned by visarga

Sun, 25 Dec 2022 05:08:51 GMT

> “LLMs can’t even do addition” > We show that they CAN add! To teach algos to LLMs, the trick is to describe the algo in enough detail so that there is no room for misinterpretation Algorithmic prompting paper https://arxiv.org/pdf/2211.09066.pdf

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

Teaching Algorithmic Reasoning via In-context Learning — HN submission

Wed, 07 Dec 2022 18:36:38 GMT

Teaching Algorithmic Reasoning via In-Context Learning

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