What Does the Rank Buy? A Spectral and Distributional Analysis of Low-Rank Adapt

arXiv:2604.27796v1 Announce Type: new Abstract: Exponential growth in the scale of modern foundation models has led to the widespread adoption of Low-Rank Adaptation (LoRA) as a

arXiv:2602.22911v4 Announce Type: replace-cross Abstract: Low-Rank Adaptation (LoRA) dominates parameter-efficient fine-tuning (PEFT). However, it faces a critical ``linear

arXiv:2609.10656v1 Announce Type: new Abstract: Selecting LoRA rank for diffusion fine-tuning requires balancing quality and compute cost. We present a controlled study on

arXiv:2609.12327v1 Announce Type: new Abstract: Reservoir computing (RC) trains only a linear readout over a fixed recurrent layer, making it fast and data-efficient for online

arXiv:2605.04115v1 Announce Type: cross Abstract: Learning in neural systems arises from synaptic changes that reshape the representations underlying behavior. While low-rank

<!-- SC_OFF --><div class="md"><p>While doing QLoRA fine-tuning (using Unsloth), increasing LoRA rank from 8,16 → 64 often doesn’t improve performance.</p> <p>It feels like it

arXiv:2609.06403v1 Announce Type: new Abstract: Test-time scaling produces cohorts of reasoning rollouts, yet there is no standard label-free account of how their internal