s1: Simple test-time scaling · HackerPapers discussions

s1: Simple test-time scaling — mentioned by embedding-shape

Mon, 20 Jul 2026 21:21:28 GMT

> My favorite trick for controlling the reasoning level is the hack First described here I think, specifically for Qwen models: https://arxiv.org/abs/2501.19393, although it seems unclear how much it actually improves the final response unless you create a new fine-tuned model with this in mind, and also if it's possible to replicate for other models. Still, got inspired and gonna give it a try with DiffusionGemma although slightly differently

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

s1: Simple test-time scaling — mentioned by mxwsn

Sun, 12 Jul 2026 20:01:49 GMT

This is the same reasoning behind why Yann Lecun thought test-time scaling would not work for LLMs: compounding error. Instead, the more tokens LLMs use, the better their performance on many tasks. LLMs can self-correct, evidenced by the power of getting models to question themselves by emitting "Wait," in S1. https://arxiv.org/abs/2501.19393

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

s1: Simple test-time scaling — mentioned by kgeist

Wed, 01 Apr 2026 05:34:48 GMT

>One theory is that the knowledge required to solve the task is already stored in the parameters of the model, and only the style has to change for task success >In particular, learning to generate longer outputs may be possible in few parameters Reminded me of: https://arxiv.org/abs/2501.19393 >we develop budget forcing to control test-time compute by forcefully terminating the model’s thinking process or lengthening it by appending “Wait” multiple times to the model’s generation when it tries to end. This can lead the model to double-check its answer, often fixing incorrect reasoning steps…

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

s1: Simple test-time scaling — HN submission

Wed, 18 Jun 2025 21:06:55 GMT

S1: Simple Test-Time Scaling

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

s1: Simple test-time scaling — mentioned by mangoman

Wed, 30 Apr 2025 00:24:12 GMT

a paper with a similar idea on scaling test time reasoning, this is sorta how all the thinking models work under the hood. https://arxiv.org/abs/2501.19393

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

s1: Simple test-time scaling — mentioned by computerex

Sun, 23 Mar 2025 17:56:48 GMT

I don’t think you have an accurate understanding of how LLMs work. https://arxiv.org/abs/2501.19393 These tokens DO extend the thinking time. We are talking about causal autoregressive language models, and so these tokens can be used to guide the generation.

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

s1: Simple test-time scaling — mentioned by lelag

Thu, 06 Mar 2025 15:51:34 GMT

If that's an issue, there's a workaround using structure generation to force it to output a </thiking> token after some threshold and force it to write the final answer. It's a method used to control thinking token generation showcased in this paper: https://arxiv.org/abs/2501.19393

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

s1: Simple test-time scaling — HN submission

Mon, 10 Feb 2025 21:13:00 GMT

s1: Simple Test-Time Scaling

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

s1: Simple test-time scaling — mentioned by armcat

Sun, 09 Feb 2025 19:47:19 GMT

Yes, the authors explicitly highlighted those two points in the abstract, in terms of them being the elicitation threshold for complex reasoning, namely, an extremely complete pre-trained foundation model, and a set of extremely high quality examples post-training. To your question on finetuning on the initial 10 million pool - intuitively, it would require tremendous amount of finetuning data to move the needle - you really won't be able to move the gradients much with just 817 examples, that initial pool is effectively enforcing pretty rigid regularization. There is now an increasing…

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

s1: Simple test-time scaling — mentioned by psytrancefan

Sun, 09 Feb 2025 13:29:44 GMT

Thanks. I had not seen the wait paper yet. That is crazy. s1: Simple test-time scaling https://arxiv.org/pdf/2501.19393

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

s1: Simple test-time scaling — HN submission

Sat, 08 Feb 2025 01:23:41 GMT

Test-time scaling new approach: extra test-time compute improves LLM reasoning

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

s1: Simple test-time scaling — mentioned by wrenky

Thu, 06 Feb 2025 16:32:35 GMT

nope you have the article right. "if you copy an existing model, you can get a pretty comparable model!" Really, [what the underlying paper](https://arxiv.org/abs/2501.19393) shows is how much you can improve existing models with several of the techniques released by openai recently (extending thought processes via "wait") It shows we can get a lot of mileage out of that technique.

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

s1: Simple test-time scaling — mentioned by ozgune

Thu, 06 Feb 2025 12:28:00 GMT

Agreed. Here are three things that I find surreal about the s1 paper. (1) The abstract changed how I thought about this domain (advanced reasoning models). The only other paper that did that for me was the "Memory Resource Management in VMware ESX Server". And that paper got published 23 years ago. (2) The model, data, and code are open source at https://github.com/simplescaling/s1. With this, you can start training your own advanced reasoning models. All you need is a thousand well-curated questions with reasoning steps. (3) More than half the references in the paper are from 2024 and Jan…

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

s1: Simple test-time scaling — mentioned by bberenberg

Wed, 05 Feb 2025 12:38:05 GMT

In case you’re not sure what S1 is, here is the original paper: https://arxiv.org/html/2501.19393v1

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

s1: Simple test-time scaling — mentioned by xena

Tue, 04 Feb 2025 17:26:19 GMT

The file explains it a bit, but my blogpost https://xeiaso.net/notes/2025/s1-simple-test-time-scaling/ could probably be better explained. I'll write out more but just for you I'll summarize what I'm gonna end up writing up. AI models fundamentally work on the basis of "given what's before, what comes next?" When you pass messages to an API like: [ { "role": "system", content": "You are an expert in selling propane and propane accessories. Whenever someone talks about anything that isn't propane, steer them back." }, { "role": "user", "content": "What should I use to cook food on my grill?"…

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