AI Paper Atlas

Research map / Continual learning / Online and replay-based

Exemplar-Free Continual Learning for State Space Models

Isaac Ning Lee, Leila Mahmoodi, Trung Le, Mehrtash Harandi

CVPR 2026 · Continual learning › Online and replay-based · approach: Unified and unsupervised continual frameworks · 0 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
integrate Inf-SSM into existing CIL baselines (ER, LUCIR, X-DER, L2P-R, CLFD)CIL baselines on Vim-small · ImageNet-R, CIFAR-100, Caltech-256 (5 and 10 tasks)AA and FM: AA +8.31% and FM -9.36% on averageImproved
add Inf-SSM to X-DERX-DER · ImageNet-R, CIFAR-100, Caltech-256AA and FM: AA +19.61%, FM -23.16% on averageImproved
add Inf-SSM to LUCIRLUCIR · ImageNet-R (5 tasks)AIA: 60.97 -> 59.53Hurt
add Inf-SSM to ERER · ImageNet-R (5 tasks)AA: 30.91 -> 31.26No clear effect
regularize observability subspace (A,C) with Inf-SSM instead of EWC/SI/MAS/LwF on (A,C)Vim-small EFCIL · ImageNet-R, CIFAR-100, Caltech-256FM and AA: FM -40.18%, AA +21.73% on averageImproved
use Inf-SSM (A,C only) instead of baselines regularizing A,B,CVim-small EFCIL · ImageNet-R, CIFAR-100, Caltech-256FM and AA: FM -14.56%, AA +6.79% on averageImproved
use diagonal-A closed-form Sylvester solution instead of Bartels-Stewart (ablation)Inf-SSM · Vim-small, n=16FLOPS: 25n^3 -> 4n^2, about 100x fewer FLOPSImproved
regularize more Vim blocks with Inf-SSM (ablation)Inf-SSM · Vim-smallperformanceImproved
add Frobenius regularization of B (Inf-SSM+) (ablation)Inf-SSM · CL benchmarksAA, AIA, FMNo clear effect
use Inf-SSM loss instead of X-DER lossCL loss cost · Vim-small, A40 GPU, per batchlatency: 1.2534 s -> 0.0960 sImproved
replace EWC with Inf-SSM lossregularizer cost · Vim-small, A40 GPUper-batch latency: 0.0095 -> 0.0960 s per batch, but no 181.5038 s per-task Fisher passHurt
Inf-SSM benefit grows with number of tasks (10 vs 5 tasks)Inf-SSM integrated with baselines · Vim-smallAA and FMImproved

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