Using an LLM fine-tuned to translate text between 11 emotions, the authors neutralize sarcastic passages for RAG question answering, improving accuracy on fully sarcastic context by about 3 percentage points.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Reading with Intent -- Neutralizing Intent
Using an LLM fine-tuned to translate text between 11 emotions, the authors neutralize sarcastic passages for RAG question answering, improving accuracy on fully sarcastic context by about 3 percentage points.