less is more: recursive reasoning with tiny networks, by a. jolicoeur-martineau (2025)

This commit is contained in:
Marina von Steinkirch, PhD 2025-10-23 17:12:34 -05:00 committed by GitHub
parent eb01bb0421
commit 1c631cf9ca
No known key found for this signature in database
GPG key ID: B5690EEEBB952194

8
trm/README.md Normal file
View file

@ -0,0 +1,8 @@
## tiny recursive models
<br>
* **[less is more: recursive reasoning with tiny networks, by a. jolicoeur-martineau (2025)](www.alphaxiv.org/pdf/2510.04871)**
* - much simpler recursive reasoning approach that achieves significantly higher generalization than HRM on a net with only 2 layers.
- with 7M parameters: 45% test-accuracy on ARC-AGI1, 8% on ARC-AGI-2 (higher than most LLMs with > 0.01% of the parameters).
- *"currently, recursive reasoning models such as HRM and TRM are supervised learning methods rather than generative models. this means that given an input question, they can only provide a single deterministic answer. in many settings, multiple answers exist for a question. thus, it would be interesting to extend TRM to generative tasks."*