EPISODE · Oct 9, 2025 · 21 MIN
Less is More: Recursive Reasoning with Tiny Networks
from Daily Paper Cast · host Jingwen Liang, Gengyu Wang
🤗 Upvotes: 89 | cs.LG, cs.AI Authors: Alexia Jolicoeur-Martineau Title: Less is More: Recursive Reasoning with Tiny Networks Arxiv: http://arxiv.org/abs/2510.04871v1 Abstract: Hierarchical Reasoning Model (HRM) is a novel approach using two small neural networks recursing at different frequencies. This biologically inspired method beats Large Language models (LLMs) on hard puzzle tasks such as Sudoku, Maze, and ARC-AGI while trained with small models (27M parameters) on small data (around 1000 examples). HRM holds great promise for solving hard problems with small networks, but it is not yet well understood and may be suboptimal. We propose Tiny Recursive Model (TRM), a much simpler recursive reasoning approach that achieves significantly higher generalization than HRM, while using a single tiny network with only 2 layers. With only 7M parameters, TRM obtains 45% test-accuracy on ARC-AGI-1 and 8% on ARC-AGI-2, higher than most LLMs (e.g., Deepseek R1, o3-mini, Gemini 2.5 Pro) with less than 0.01% of the parameters.
What this episode covers
🤗 Upvotes: 89 | cs.LG, cs.AI Authors: Alexia Jolicoeur-Martineau Title: Less is More: Recursive Reasoning with Tiny Networks Arxiv: http://arxiv.org/abs/2510.04871v1 Abstract: Hierarchical Reasoning Model (HRM) is a novel approach using two small neural networks recursing at different frequencies. This biologically inspired method beats Large Language models (LLMs) on hard puzzle tasks such as Sudoku, Maze, and ARC-AGI while trained with small models (27M parameters) on small data (around 1000 examples). HRM holds great promise for solving hard problems with small networks, but it is not yet well understood and may be suboptimal. We propose Tiny Recursive Model (TRM), a much simpler recursive reasoning approach that achieves significantly higher generalization than HRM, while using a single tiny network with only 2 layers. With only 7M parameters, TRM obtains 45% test-accuracy on ARC-AGI-1 and 8% on ARC-AGI-2, higher than most LLMs (e.g., Deepseek R1, o3-mini, Gemini 2.5 Pro) with less than 0.01% of the parameters.
NOW PLAYING
Less is More: Recursive Reasoning with Tiny Networks
No transcript for this episode yet
Similar Episodes
No similar episodes found.