An auto-review pipeline using multiple ASR transcript alternatives and LLM-generated pseudo-labels improves field extraction accuracy on healthcare benefit calls, but gains are inconsistent outside a fine-tuned model.
group number is 123 456 789
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
-
Auto Review: Second Stage Error Detection for Highly Accurate Information Extraction from Phone Conversations
An auto-review pipeline using multiple ASR transcript alternatives and LLM-generated pseudo-labels improves field extraction accuracy on healthcare benefit calls, but gains are inconsistent outside a fine-tuned model.