List your company

Health Misinformation Detection on HealthFC

84Macro F1 ScoreiPOE-h
36.1648.586173.42Nov 29, 2025Dec 27, 2025Jan 24, 2026Feb 22, 2026Mar 22, 2026Apr 19, 2026May 18, 2026

Evaluation results

MethodLinks
iPOE-hModel=Qwen3-4B, Protocol=Zero-shot2026.0584——
iPOE-lb-llmModel=Qwen3-4B, Protocol=Zero-shot2026.0584——
iPOE-hModel=LLama3-8B, Protocol=Zero-shot2026.0584——
iPOE-lb-hModel=Qwen3-4B, Protocol=Zero-shot2026.0583——
Multi-Agent DebateEvaluation stages=1st stage + 2nd stage2025.1182.482.182.7
StepByStep2025.118172.691.6
iPOE-llmModel=Qwen3-4B, Protocol=Zero-shot2026.0580——
iPOE-hModel=Qwen3-30B, Protocol=Zero-shot2026.0580——
iPOE-llmModel=LLama3-8B, Protocol=Zero-shot2026.0579——
iPOE-llmModel=Qwen3-30B, Protocol=Zero-shot2026.0579——
iPOE-lb-llmModel=Qwen3-30B, Protocol=Zero-shot2026.0579——
WebAgent2025.1178.17878.3
iPOE-lb-llmModel=LLama3-8B, Protocol=Zero-shot2026.0578——
Rand-hModel=Qwen3-30B, Protocol=Zero-shot2026.0577——
iPOE-lb-hModel=Qwen3-30B, Protocol=Zero-shot2026.0577——
iPOE-lb-hModel=LLama3-8B, Protocol=Zero-shot2026.0576——
Rand-llmModel=Qwen3-30B, Protocol=Zero-shot2026.0576——
Agreement Score PredictionEvaluation stages=1st stage only2025.1174.376.973.4
VanillaModel=Qwen3-30B, Protocol=Zero-shot2026.0571——
Rand-llmModel=Qwen3-4B, Protocol=Zero-shot2026.0565——
Rand-llmModel=LLama3-8B, Protocol=Zero-shot2026.0563——
VanillaModel=Qwen3-4B, Protocol=Zero-shot2026.0560——
VanillaModel=LLama3-8B, Protocol=Zero-shot2026.0548——
Rand-hModel=Qwen3-4B, Protocol=Zero-shot2026.0546——
Rand-hModel=LLama3-8B, Protocol=Zero-shot2026.0538——