AlphaWiSE: Adaptive Weight Interpolation for Continual Multimodal Representation Learning
Multimodal models such as CLIP learn a shared embedding space for cross-modal retrieval, but continual adaptation to sequentially arriving data can disrupt the cross-modal alignment acquired from earlier phases. Conventional continual-learning methods return a single checkpoint, which commits every retrieval direction to the same stability-plasticity trade-off. We propose AlphaWiSE, a post-hoc weight-space interpolation method that composes two frozen source checkpoints. For each aligned parameter tensor identified by its checkpoint key, AlphaWiSE fits one scalar interpolation coefficient shar
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- FuzzyOverlapping authors or contributors · 62%firecrawl/firecrawl →
“Shared author/contributor keys: jain”
- FuzzyOverlapping authors or contributors · 62%modular/modular →
“Shared author/contributor keys: liu”
- FuzzySimilar title/name (fuzzy) · 59%aymericdamien/TopDeepLearning →
“Fuzzy title match (0.73): “AlphaWiSE: Adaptive Weight Interpolation for Continual Multi” ≈ “aymericdamien/TopDeepLearning””
- LinkedLinked via arxiv author · 85%Sarthak Jain →
“AlphaWiSE: Adaptive Weight Interpolation for Continual Multimodal Representation Learning”
- LinkedLinked via arxiv author · 85%Qiran Hu →
“AlphaWiSE: Adaptive Weight Interpolation for Continual Multimodal Representation Learning”
- LinkedLinked via arxiv author · 85%Zhen Zhu →
“AlphaWiSE: Adaptive Weight Interpolation for Continual Multimodal Representation Learning”
- LinkedLinked via arxiv author · 85%Yaoyao Liu →
“AlphaWiSE: Adaptive Weight Interpolation for Continual Multimodal Representation Learning”