Invariant semantic features in language models are characterized as geometric subspaces in latent space, separated via contrastive discovery and used for model attribution.
Universal adversarial triggers for attacking and analyzing NLP,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Invariant Features in Language Models: Geometric Characterization and Model Attribution
Invariant semantic features in language models are characterized as geometric subspaces in latent space, separated via contrastive discovery and used for model attribution.