Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

ReasoningGym: Reasoning Environments for RL with Verifiable Rewards

arxiv.org

Post reply on HN
Print view View original

11–20 of 34 posts

Re: ReasoningGym: Reasoning Environments for RL with Verifiable Rewards

#12

Post by CuriouslyC »

phh wrote: post #4

Cool cool. I'm a bit put off by calling it "reasoning" /"thought". These RL targets can be achieved without "thinking" model but still cool. Gotta love the brainfuck task. I personally think that Gemini 2.5 Pro's superiority comes from having hundreds or thousands RL tasks (without any proof whatsoever, so rather a feeling). So I've been wanting a "RL Zoo" for quite a while. I hope this project won't be a one-off and…

Gemini 2.5 Pro's superiority is IMO largely driven by their long context support and training methodology. Compare Gemini as a beta reader for a 100k token book with GPT4.1 or Claude 4, and it becomes quite clear how much more effectively it can reason across its context than other comparable models. This also makes it much better for architecting new features into a system, since you can load a lot of the current system into the context and it'll conform to existing styles and architecture patterns more closely.

Re: ReasoningGym: Reasoning Environments for RL with Verifiable Rewards

#13

Post by phh »

olliestanley wrote: post #6
phh wrote: post #4

Cool cool. I'm a bit put off by calling it "reasoning" /"thought". These RL targets can be achieved without "thinking" model but still cool. Gotta love the brainfuck task. I personally think that Gemini 2.5 Pro's superiority comes from having hundreds or thousands RL tasks (without any proof whatsoever, so rather a feeling). So I've been wanting a "RL Zoo" for quite a while. I hope this project won't be a one-off and…

We definitely plan to maintain the project for as long as there is interest in it. If you have ideas for new tasks, we'd always welcome contributions!

Thanks for the answer! As a toy project I implemented wikiracing with trl. I'll probably try to PR that to your gym. (can't say that I managed to improve score with it though)

Re: ReasoningGym: Reasoning Environments for RL with Verifiable Rewards

#16

Post by yorwba »

t55 wrote: post #8
ninakostoska wrote: post #7

Cool to see NVIDIA’s most recent reasoning model [1] already uses Reasoning Gymas a large part of their data mixture [1] https://arxiv.org/abs/2505.24864

> prolonged RL training can uncover novel reasoning strategies that are inaccessible to base models, even under extensive sampling does this mean that previous RL papers claiming the opposite were possibly bottlenecked by small datasets?

No, they do not point to any specific examples of novel reasoning strategies that were uncovered, nor is their sampling that extensive (at most 256 samples vs the 2048 used in https://limit-of-rlvr.github.io/ ).

Re: ReasoningGym: Reasoning Environments for RL with Verifiable Rewards

#17

Post by jacob019 »

CuriouslyC wrote: post #12
phh wrote: post #4

Cool cool. I'm a bit put off by calling it "reasoning" /"thought". These RL targets can be achieved without "thinking" model but still cool. Gotta love the brainfuck task. I personally think that Gemini 2.5 Pro's superiority comes from having hundreds or thousands RL tasks (without any proof whatsoever, so rather a feeling). So I've been wanting a "RL Zoo" for quite a while. I hope this project won't be a one-off and…

Gemini 2.5 Pro's superiority is IMO largely driven by their long context support and training methodology. Compare Gemini as a beta reader for a 100k token book with GPT4.1 or Claude 4, and it becomes quite clear how much more effectively it can reason across its context than other comparable models. This also makes it much better for architecting new features into a system, since you can load a lot of the current sy…

Agreed, 2.5 flash too. I analyze a large json document of metrics for pricing decisions. Typically around 200k, occtionallly up to 1M, Gemini 2.5 significantly outperforms for my task. It isn't 100%, but role playing gets close. I suppose that's a form of inference time compute.
Post reply on HN