AI Paper Atlas

Research map / Continual learning / Continual fine-tuning of LLMs

COBRA: A Continual Learning Approach to Vision-Brain Understanding

Xuan-Bac Nguyen, Manuel Serna-Aguilera, Arabinda K. Choudhary, Pawan Sinha, Xin Li, Ky Luu

IJCV 2026 · Continual learning › Continual fine-tuning of LLMs · approach: Mixture-of-experts for multimodal continual tuning · 4 citations (OpenAlex)

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
add COBRA continual learning (SC + PSS + MRIFormer) versus no continual learningCOBRA · NSD rehearsal-free reconstruction (4 steps, all)SSIM / CLIP: 0.207 -> 0.328 SSIM, 59.8 -> 93.2 CLIPImproved
use COBRA instead of PLOP in rehearsal-free settingcontinual learning method · NSD rehearsal-free reconstruction (4 steps, all)SSIM / CLIP: 0.288 -> 0.328 SSIM, 76.0 -> 93.2 CLIPImproved
combine COBRA with PLOPCOBRA · NSD rehearsal-free reconstruction (4 steps, all)SSIM / CLIP: 0.328 -> 0.338 SSIM, 93.2 -> 94.2 CLIPImproved
combine COBRA with PLOPCOBRA · NSD rehearsal-free reconstruction (2 steps, all)SSIM / CLIP: 0.329 -> 0.333 SSIM, 92.9 -> 93.7 CLIPImproved
use COBRA instead of LwF in 2-step rehearsal-free settingcontinual learning method · NSD rehearsal-free reconstruction (2 steps, all)SSIM / CLIP: 0.249 -> 0.329 SSIM, 64.3 -> 92.9 CLIPImproved
increase rehearsal buffer size from 500 to 4000 samples per subject (ablation)COBRA · NSD rehearsal-based reconstruction (4 steps, all)SSIM / CLIP: 0.334 -> 0.343 SSIM, 93.6 -> 95.8 CLIPImproved
use COBRA instead of Co2L in rehearsal-based settingcontinual learning method · NSD rehearsal-based reconstruction (4 steps, buffer 500, all)SSIM / CLIP: 0.300 -> 0.334 SSIM, 83.1 -> 93.6 CLIPImproved
use full-brain fMRI signals instead of partial ROI signals as input (ablation)COBRA · NSD reconstructionSSIM / CLIP: 3% higher SSIM and 2% higher CLIPImproved
use MRIFormer transformer encoder-decoder instead of linear layers (ablation)COBRA · NSD reconstructionSSIM / CLIP: approximately 3% in both SSIM and CLIPImproved
set PSS prompt top-k to 30 instead of larger or smaller values (ablation)PSS module · NSD reconstructionSSIM / CLIPImproved
select subject-specific tokens by prompt-based PSS instead of attention pooling (ablation)PSS module · Subject identification accuracyaccuracy: 82.8 -> 98.5Improved
select tokens by prompt-based PSS instead of random (ablation)PSS module · Subject identification accuracyaccuracy: 21.6 -> 98.5Improved

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