AI Paper Atlas

Research map / Learning theory and algorithms / Optimal transport

Representational Alignment Across Model Layers and Brain Regions with Hierarchical Optimal Transport

Shaan Shah, Meenakshi Khosla

ICLR 2026 · Learning theory and algorithms › Optimal transport · approach: Neural optimal transport

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use global soft layer coupling (MOT) instead of independent pairwise best-layer matchingMOT · LLM alignment (Llama-3.2 1B vs 3B)held-out reconstruction correlation: 0.505 -> 0.558Improved
use MOT instead of pairwise best OTMOT · NSD cortex alignment (Subject A vs B)held-out reconstruction correlation: 0.245 -> 0.244No clear effect
use vanilla MOT instead of pairwise best OTMOT · Vision models (ViT-MAE Base vs Huge)held-out reconstruction correlation: 0.417 -> 0.149Hurt
use vanilla MOT instead of pairwise best OTMOT · Vision models (ViT-MAE Base vs DINOv2 Large)held-out reconstruction correlation: 0.624 -> 0.577Improved
add rotation (MOT+R) to MOT (ablation)MOT · Vision models (DINOv2 Small vs Large)held-out reconstruction correlation: 0.353 -> 0.778Improved
use MOT+R instead of rotation-aware pairwise bestMOT+R · Vision models (DINOv2 Small vs Giant)held-out reconstruction correlation: 0.418 -> 0.790Improved
keep soft layer coupling instead of top-1 hard assignment (ablation)MOT · LLM alignment (Llama-3.2 1B vs 3B)held-out reconstruction correlation: 0.502 -> 0.558Improved
optimize layer coupling instead of random permutation of P (ablation)MOT · LLM alignment (Llama-3.2 1B vs 3B)held-out reconstruction correlation: 0.510 -> 0.558Improved
optimize layer coupling instead of random permutation of P (ablation)MOT · NSD cortex alignment (Subject A vs B)held-out reconstruction correlation: 0.135 -> 0.244Improved
use layer-wise MOT instead of flattening all neurons into one global OT (ablation)MOT · NSD cortex alignment (Subject A vs B)held-out reconstruction correlation: 0.184 -> 0.244Improved
use MOT instead of global linear predictivityMOT · NSD cortex alignment (Subject A vs B)held-out reconstruction correlation: 0.305 -> 0.244Hurt
use MOT instead of pairwise linear predictivityMOT · LLM alignment (Llama-3.2 1B vs 3B)held-out reconstruction correlation: 0.737 -> 0.558Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy