Hybrid human-LLM crowds with locally weighted aggregation outperform human-only and LLM-only groups on both accuracy and bias reduction for headline verification.
Towards controllable biases in language generation
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
-
Wisdom from Diversity: Bias Mitigation Through Hybrid Human-LLM Crowds
Hybrid human-LLM crowds with locally weighted aggregation outperform human-only and LLM-only groups on both accuracy and bias reduction for headline verification.