DySem selects dynamic semantic dimensions from LLMs via multilingual consensus and computes similarity over text-specific shared subsets, claiming better performance at lower dimensionality than fixed last-layer baselines.
InProceedings of the 8th Interna- tional Workshop on Semantic Evaluation (SemEval 2014), pages 81–91, Dublin, Ireland
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
DySem: Uncovering Dynamic Semantic Components of Large Language Models for Calculating Semantic Textual Similarity
DySem selects dynamic semantic dimensions from LLMs via multilingual consensus and computes similarity over text-specific shared subsets, claiming better performance at lower dimensionality than fixed last-layer baselines.