DiaFORGE is a disambiguation-centric fine-tuning pipeline that boosts LLM tool-calling success by 27-49 percentage points on a dynamic benchmark compared to GPT-4o and Claude-3.5-Sonnet.
- **Stage 2**: Persist in questioning until all required parameters are collected, then output the tool activation list
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Disambiguation-Centric Finetuning Makes Enterprise Tool-Calling LLMs More Realistic and Less Risky
DiaFORGE is a disambiguation-centric fine-tuning pipeline that boosts LLM tool-calling success by 27-49 percentage points on a dynamic benchmark compared to GPT-4o and Claude-3.5-Sonnet.