SALU, a multi-task fine-tuning and confidence-guided RLHF method, reduces hallucinated answers on unanswerable Chinese CIR questions to 1.3 percent on the authors' private dataset.
How does bert answer questions? a layer-wise analysis of transformer representations,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Harnessing RLHF for Robust Unanswerability Recognition and Trustworthy Response Generation in LLMs
SALU, a multi-task fine-tuning and confidence-guided RLHF method, reduces hallucinated answers on unanswerable Chinese CIR questions to 1.3 percent on the authors' private dataset.