Research group finds reasoning models produce fractals when solving hard problems
wgilpin0 · x · 2026-09-09
The team discovered that reasoning models produce fractals when solving hard problems. They parametrize problem + weights as a dynamical system with solutions as fixed points, varying each replicate's initial latent state across a random 2D slice. Nonlinear dynamics lets them probe recurrent depth models on Sudoku, math, and even ARC-AGI, and explains 'overthinking': similar prompts can yield 10x differences in reasoning length.
Related event: Fractal basins and transient chaos explain why reasoning models overthink(13 posts)→
More from Research
- Strogatz: a Navier-Stokes singularity wouldn't invalidate the model, just end the continuum approximation — stevenstrogatz · 2026-09-09
- Vortex stretching vs viscosity: the battle inside Navier-Stokes that could breed a singularity — stevenstrogatz · 2026-09-09
- Vortex Tubes Get Thinner and Spin Faster: Strogatz on Vorticity — stevenstrogatz · 2026-09-09
- Mathematician Steven Strogatz used ChatGPT on his walk to grasp the claimed Navier-Stokes singular solution — stevenstrogatz · 2026-09-09
- 3D Kakeya conjecture fully formalized in Lean 4 with no sorry, aided by ByteDance Seed AI4Math — AlexKontorovich · 2026-09-09
- WaveNet Turns 10: Author Traces Speech Synthesis from Waveforms to Multimodal LLMs — heiga_zen · 2026-09-09