In an exact-fit linear regime with i.i.d. tasks from distribution Π, forgetting obeys a recursive spectral operator whose asymptotic convergence rate is governed by geometric properties of Π.
https: //arxiv.org/abs/2004.07211
6 Pith papers cite this work, alongside 90 external citations. Polarity classification is still indexing.
6
Pith papers citing it
90
external citations · external index
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 6roles
background 1polarities
background 1representative citing papers
A deterministic episodic-to-semantic consolidation function with a structural lemma proving identity invariance, demonstrated in synthetic experiments on an embodied service agent.
CVLC fuses calibrated vision prototypes with LLM-generated language prototypes and applies dual coalescent projection plus latent space reservation to enable few-shot adaptation across sequential domains, reporting up to 16% gains over prior methods.
AIM applies modality-specific masks to balance stability and plasticity in asymmetric VLMs, achieving SOTA average performance and reduced forgetting on continual VQA v2 and GQA while preserving generalization to novel compositions.
A plug-and-play KL regularizer that masks the target token and renormalizes probabilities to improve the learning-forgetting trade-off in LoRA adaptation of LLMs.
BRAIN uses bias-mitigation continual learning with a new de-bias contrastive loss and angular forgetting mitigation to achieve SOTA performance on vision-brain understanding benchmarks despite brain signal inconsistencies across sessions.
citing papers explorer
-
From Order to Distribution: A Spectral Characterization of Forgetting in Continual Learning
In an exact-fit linear regime with i.i.d. tasks from distribution Π, forgetting obeys a recursive spectral operator whose asymptotic convergence rate is governed by geometric properties of Π.
-
Episodic-to-Semantic Consolidation Without Identity Drift
A deterministic episodic-to-semantic consolidation function with a structural lemma proving identity invariance, demonstrated in synthetic experiments on an embodied service agent.
-
Few-Shot Domain Incremental Learning via Continual Vision-Language Consolidation
CVLC fuses calibrated vision prototypes with LLM-generated language prototypes and applies dual coalescent projection plus latent space reservation to enable few-shot adaptation across sequential domains, reporting up to 16% gains over prior methods.
-
AIM: Asymmetric Information Masking for Visual Question Answering Continual Learning
AIM applies modality-specific masks to balance stability and plasticity in asymmetric VLMs, achieving SOTA average performance and reduced forgetting on continual VQA v2 and GQA while preserving generalization to novel compositions.
-
Mask the Target: A Plug-and-Play Regularizer Against LoRA Forgetting
A plug-and-play KL regularizer that masks the target token and renormalizes probabilities to improve the learning-forgetting trade-off in LoRA adaptation of LLMs.
-
BRAIN: Bias-Mitigation Continual Learning Approach to Vision-Brain Understanding
BRAIN uses bias-mitigation continual learning with a new de-bias contrastive loss and angular forgetting mitigation to achieve SOTA performance on vision-brain understanding benchmarks despite brain signal inconsistencies across sessions.