LLMs often misalign their self-perceived need for tools with true need and utility, but lightweight estimators trained on hidden states can improve tool-calling decisions and task performance across multiple models and tasks.
Adaptive Retrieval Without Self-Knowledge? Bringing Uncertainty Back Home
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
To Call or Not to Call: A Framework to Assess and Optimize LLM Tool Calling
LLMs often misalign their self-perceived need for tools with true need and utility, but lightweight estimators trained on hidden states can improve tool-calling decisions and task performance across multiple models and tasks.