The paper proposes information scope as a new interpretability axis for SAE features in CLIP and introduces the Contextual Dependency Score to separate local from global scope features, showing they influence model predictions differently.
De- coding dense embeddings: Sparse autoencoders for inter- preting and discretizing dense retrieval
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3representative citing papers
Xetrieval enriches sentence embeddings with a single-pass reasoning internalizer and decomposes the result into sparse interpretable features whose overlaps explain individual dense-retrieval decisions.
Replacing SPLADE's MLM vocabulary with SAE-learned semantic concepts achieves comparable retrieval performance with improved efficiency.
citing papers explorer
-
Beyond Semantics: Disentangling Information Scope in Sparse Autoencoders for CLIP
The paper proposes information scope as a new interpretability axis for SAE features in CLIP and introduces the Contextual Dependency Score to separate local from global scope features, showing they influence model predictions differently.
-
Xetrieval: Mechanistically Explaining Dense Retrieval
Xetrieval enriches sentence embeddings with a single-pass reasoning internalizer and decomposes the result into sparse interpretable features whose overlaps explain individual dense-retrieval decisions.
-
From Tokens to Concepts: Leveraging SAE for SPLADE
Replacing SPLADE's MLM vocabulary with SAE-learned semantic concepts achieves comparable retrieval performance with improved efficiency.