List your company

Fact Verification on FEVER (Accuracy and Tokens)

78AccuracyClean Baseline
54.0860.2966.572.71Dec 16, 2025

Evaluation results

MethodLinks
Clean BaselineBackbone=Qwen3-32B, Condition=Clean, Architecture=Tree of Thoughts2025.127811,500
Clean BaselineBackbone=Qwen3-32B, Condition=Clean, Architecture=Reflexion2025.12753,200
Clean BaselineBackbone=Qwen3-32B, Condition=Clean, Architecture=ReAct2025.12721,800
Meta-HasteBackbone=Qwen3-32B, Condition=Baseline Attack, Architecture=Tree of Thoughts2025.126511,200
Meta-HasteBackbone=Qwen3-32B, Condition=Baseline Attack, Architecture=Reflexion2025.12633,000
GSI-HasteBackbone=Qwen3-32B, Condition=Ours, Architecture=Tree of Thoughts2025.126210,800
Meta-HasteBackbone=Qwen3-32B, Condition=Baseline Attack, Architecture=ReAct2025.12611,700
GSI-HasteBackbone=Qwen3-32B, Condition=Ours, Architecture=Reflexion2025.12582,900
GSI-HasteBackbone=Qwen3-32B, Condition=Ours, Architecture=ReAct2025.12551,600