← Back to the feed
Space & aviationResearch findings

Spectral and Subspace Dynamics in LoRA Across SFT and RL

Original published 2026.11.06 03:17 KSTAdded to Blackdeck 2026.11.09 14:07 KST

What has changed?

In RL, the spectrum concentrates at all learning rates, while SFT expands it at moderate rates; RL also rotates singular subspaces faster than SFT at low rates.

Key points

  1. SFT and RL show distinct spectral dynamics in LoRA updates.
  2. RL concentrates the spectrum, while SFT expands it at moderate learning rates.
  3. RL rotates singular subspaces faster than SFT at low learning rates.

Connect it to your work

Group-mean baseline subtraction is identified as the root cause of these dynamics.

What remains unproven

This research note is based on an abstract; check the original for test conditions and application scope.

Originals used for this note

Links open the original publisher in a new tab.

  1. arxivSpectral and Subspace Dynamics of LoRA in Supervised Fine-Tuning and Reinforcement Learning