Shifts in an LLM's hidden-state confidence, before and after a retrieved context, are used as a preference signal to fine-tune a reranker and to trigger retrieval only when initial confidence is low.
Are large language models more honest in their probabilistic or verbalized confidence? InChina Conference on Information Retrieval, pages 124–135
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Rethinking LLM Parametric Knowledge as Post-retrieval Confidence for Dynamic Retrieval and Reranking
Shifts in an LLM's hidden-state confidence, before and after a retrieved context, are used as a preference signal to fine-tune a reranker and to trigger retrieval only when initial confidence is low.