PRISM forms predictions as sparse mixtures of learned prototypes trained with clustering objectives, matching dense model accuracy while enabling ~500x faster data attribution and behavior editing without finetuning.
Proceedings of the 40th International Conference on Machine Learning , series =
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
Interpreting harmful Discord messages requires integrating external knowledge and extended context, not just local message-level classification; LLMs leverage local context better than humans but still fail on coded language and community-specific references.
citing papers explorer
-
Prototype Language Models
PRISM forms predictions as sparse mixtures of learned prototypes trained with clustering objectives, matching dense model accuracy while enabling ~500x faster data attribution and behavior editing without finetuning.
-
Understanding Interpretation Difficulty in Harmful Online Communication: Insights from Cybercrime Communities
Interpreting harmful Discord messages requires integrating external knowledge and extended context, not just local message-level classification; LLMs leverage local context better than humans but still fail on coded language and community-specific references.