Less is More: Recursive Reasoning with Tiny Networks · HackerPapers discussions

Less is More: Recursive Reasoning with Tiny Networks — mentioned by calebkaiser

Mon, 19 Oct 2026 08:32:17 GMT

Sebastian Raschka has a good introductory piece on what Looped Transformers usually refer to in practice: https://magazine.sebastianraschka.com/p/gpt-6-astra-looped-t... One of the big points is that looped architectures don't necessarily remove CoT style reasoning. They definitely change it, but the model can still be prompted for this kind of output. It just spends more computation on individual tokens. I think the hope, in that particular scenario, is that the model spends less tokens on things like backtracking. But there are lots of approaches to looped transformer architectures. Some…

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

Less is More: Recursive Reasoning with Tiny Networks — mentioned by SamCoding

Thu, 23 Jul 2026 20:50:27 GMT

Referring to this paper: https://arxiv.org/pdf/2510.04871 My understanding: The Q-head predicts whether to stop early or not to reduce training times (by not always having to complete the full number of deep supervision steps). The complication: When working in batches: - Either, the Q-head is computed over the entire batch and averaged (which could early stopping of reasoning steps when most items in batch are complete but some aren't, meaning that those incomplete samples aren't trained on fully) - Or, the Q-head is computed over entire batch and samples that are complete are dropped. The…

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

Less is More: Recursive Reasoning with Tiny Networks — mentioned by whiplash451

Wed, 10 Jun 2026 11:48:48 GMT

The TRM architecture models both the problem and the solution at the same time. You might find it an interesting read. https://arxiv.org/abs/2510.04871

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

Less is More: Recursive Reasoning with Tiny Networks — mentioned by scoring1774

Thu, 05 Feb 2026 20:36:20 GMT

This has been done: https://arxiv.org/abs/2510.04871v1

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

Less is More: Recursive Reasoning with Tiny Networks — mentioned by m_ke

Sat, 22 Nov 2025 20:55:04 GMT

Yes it would require completely new hardware and most likely ditching gradient descent for alternative optimization methods, though I'm not convinced that we'd need to turn to discrete optimization. Some recent works that people might find interesting: - Evolution Strategies at the Hyperscale - https://eshyperscale.github.io/ - Introducing Nested Learning: A new ML paradigm for continual learning - https://research.google/blog/introducing-nested-learning-a-n... - Less is More: Recursive Reasoning with Tiny Networks - https://arxiv.org/abs/2510.04871 - Nemotron Elastic: Towards Efficient…

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

Less is More: Recursive Reasoning with Tiny Networks — mentioned by JSR_FDED

Fri, 14 Nov 2025 01:54:48 GMT

https://arxiv.org/abs/2510.04871

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

Less is More: Recursive Reasoning with Tiny Networks — mentioned by ayazhan

Wed, 15 Oct 2025 20:21:06 GMT

https://arxiv.org/abs/2510.04871 another recursive based model

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

Less is More: Recursive Reasoning with Tiny Networks — HN submission

Fri, 10 Oct 2025 18:43:52 GMT

Less is More: An LLM that outscores Claude Sonnet 4 while being 50.000x smaller

5 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.

Less is More: Recursive Reasoning with Tiny Networks — HN submission

Fri, 10 Oct 2025 03:19:10 GMT

Less is More: An LLM that outscores Claude Sonnet 4 while being 50.000x smaller

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

Less is More: Recursive Reasoning with Tiny Networks — HN submission

Thu, 09 Oct 2025 02:23:09 GMT

Less Is More: Recursive Reasoning with Tiny Networks

2 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.

Less is More: Recursive Reasoning with Tiny Networks — HN submission

Wed, 08 Oct 2025 14:55:44 GMT

Show HN: Recursive Reasoning with Tiny Networks in the Browser

I wanted to get a better understanding of this paper, so I asked Claude to train and generate a small sudoku solver with some extra explaining. Original paper: https://arxiv.org/abs/2510.04871 New Link including link to source in github: https://claude.ai/public/artifacts/1dde5d3c-4ad8-420f-9b62-4...

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

Less is More: Recursive Reasoning with Tiny Networks — HN submission

Wed, 08 Oct 2025 14:04:40 GMT

Samsung released a 7M model that achieved 45% on ARC-AGI-1

34 points; 12 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Less is More: Recursive Reasoning with Tiny Networks — HN submission

Tue, 07 Oct 2025 17:42:06 GMT

Less is more: Recursive reasoning with tiny networks

Paper: https://arxiv.org/abs/2510.04871, Code: https://github.com/SamsungSAILMontreal/TinyRecursiveModels

323 points; 71 discussion replies on HN. Paper and source history. A submission is not an endorsement.