New LLM-based models for fine-grained conditional probability estimation outperform prior fine-tuned and prompting methods through enhanced data creation and supervision.
- moderately likely(probability~0.7): Premises that are related to the passage but are more general, potentially requiring additional context to confirm the hypothesis
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Always Tell Me The Odds: Fine-grained Conditional Probability Estimation
New LLM-based models for fine-grained conditional probability estimation outperform prior fine-tuned and prompting methods through enhanced data creation and supervision.