AI Paper Atlas

Research map / Continual learning / Online and replay-based

Semantic Concept Steering Breaks the Explanation Drift Loop in Continual Learning

Yehonatan Elisha, Oren Barkan, Noam Koenigstein

NeurIPS 2026 · Continual learning › Online and replay-based · approach: Theoretical analysis of forgetting · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes a semantic concept steering approach to continual learning that addresses explanation drift, where models keep accuracy while shifting attention to spurious features, by fixing a flaw in pixel-level saliency regularization.

Problem

Continual learning models suffer explanation drift (attention shifting from diagnostic to spurious features) even when accuracy is maintained; existing pixel-level saliency regularization derives its supervision targets from the model's own drifting saliency maps.

Approach

Identifies a structural flaw in explanation-aware CL: pixel-level saliency regularization takes supervision targets from the model's own (already drifted) maps. Proposes replacing these with semantic concept-level steering to break the drift loop. Details beyond the abstract are not available.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace pixel-level saliency targets with concept-level targets (C3L with AttnLRP vs RRR-ViT-AttnLRP)RRR-ViT-AttnLRP · CUB (ViT-B)ACC: 78.09 -> 83.65Improved
add concept overlap loss L_concept (ablation)C3L · CUB (ViT-B)ACC: 73.74 -> 83.65Improved
add background stability loss L_stab (ablation)C3L · CUB (ViT-B)ACC: 78.42 -> 83.65Improved
add concept overlap loss L_concept (ablation)C3L · ImageNet-R (ViT-B)ACC: 77.12 -> 82.74Improved
add background stability loss L_stab (ablation)C3L · ImageNet-R (ViT-B)ACC: 78.94 -> 82.74Improved
stabilize background (cosine) rather than suppress it (ablation)C3L · ImageNet-R (ViT-B)ACC: 78.94 -> 82.74Improved
add replay buffer to semantic regularization (full C3L vs NoReplay) (ablation)C3L · ImageNet-R (ViT-B)ACC: 79.82 -> 82.74Improved
apply semantic regularization without any replay (C3L-NoReplay vs ER) (ablation)ER · ImageNet-R (ViT-B)ACC: 72.27 -> 79.82Improved
regularize replay with saved model saliency targets (FullReplay) instead of concept targets (ablation)C3L · ImageNet-R (ViT-B)ACC: 82.74 -> 80.32Hurt
use AttnLRP instead of Grad-CAM as explanation method (ablation)C3L · ImageNet-R (ViT-B)ACC: 80.92 -> 82.74Improved
reduce replay buffer from 1000 to 200 (ablation)C3L · ImageNet-R (ViT-B)ACC: 82.74 -> 80.40Hurt
use LRP/AttnLRP in RRR instead of Grad-CAMRRR-ViT · CUB (ViT-B)ACC: 76.94 -> 78.09Improved

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