TTPA teaches LLMs correct tool calls by sampling near-tie token choices, scoring them with an error taxonomy, and training with DPO.
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
-
TTPA: Token-level Tool-use Preference Alignment Training Framework with Fine-grained Evaluation
TTPA teaches LLMs correct tool calls by sampling near-tie token choices, scoring them with an error taxonomy, and training with DPO.