List your company

Question Answering on HotpotQA PIA (test)

90.2ASRVanilla
4.71226.90649.171.294Jun 4, 2025Aug 1, 2025Sep 28, 2025Nov 25, 2025Jan 22, 2026Mar 21, 2026May 19, 2026

Evaluation results

MethodLinks
VanillaLLM=Llama 2-C2025.0690.2—3.6
VanillaLLM=GPT-4o2025.0678.8—10.6
RobustRAGLLM Backbone=Llama-3.1-8B, Retrieval Model=Contriever2026.057852—
RobustRAGLLM Backbone=Qwen2.5-7B, Retrieval Model=ANCE2026.057450—
RobustRAGLLM Backbone=Qwen2.5-7B, Retrieval Model=Contriever2026.057352—
RobustRAGLLM Backbone=Llama-3.1-8B, Retrieval Model=ANCE2026.057342—
RobustRAGLLM Backbone=Llama-3.1-8B, Retrieval Model=DPR2026.057042—
VanillaLLM=Mistral-7B2025.0669—18.6
InstructRAGLLM Backbone=Mistral-7B, Retrieval Model=ANCE2026.056729—
RobustRAGLLM Backbone=Qwen2.5-7B, Retrieval Model=DPR2026.056541—
InstructRAGLLM Backbone=Mistral-7B, Retrieval Model=DPR2026.056529—
InstructRAGLLM Backbone=Qwen2.5-7B, Retrieval Model=DPR2026.056353—
InstructRAGLLM Backbone=Qwen2.5-7B, Retrieval Model=ANCE2026.056049—
ReliabilityRAGLLM Backbone=Llama-3.1-8B, Retrieval Model=DPR2026.056033—
InstructRAGLLM Backbone=Qwen2.5-7B, Retrieval Model=Contriever2026.055759—
InstructRAGLLM Backbone=Mistral-7B, Retrieval Model=Contriever2026.055740—
InstructRAGLLM Backbone=Llama-3.1-8B, Retrieval Model=ANCE2026.055350—
ReliabilityRAGLLM Backbone=Llama-3.1-8B, Retrieval Model=Contriever2026.055140—
BiRDLLM Backbone=Llama-3.1-8B, Retrieval Model=DPR2026.055064—
ReliabilityRAGLLM Backbone=Llama-3.1-8B, Retrieval Model=ANCE2026.054931—
BiRDLLM Backbone=Qwen2.5-7B, Retrieval Model=DPR2026.054854—
InstructRAGLLM Backbone=Llama-3.1-8B, Retrieval Model=Contriever2026.054442—
AV FilterLLM=Llama 2-C, alpha=∞2025.0644—17.8
VanillaRAGLLM Backbone=Llama-3.1-8B, Retrieval Model=DPR2026.054356—
ReliabilityRAGLLM Backbone=Qwen2.5-7B, Retrieval Model=ANCE2026.054233—
VanillaRAGLLM Backbone=Qwen2.5-7B, Retrieval Model=DPR2026.054255—
ReliabilityRAGLLM Backbone=Qwen2.5-7B, Retrieval Model=DPR2026.054239—
VanillaRAGLLM Backbone=Mistral-7B, Retrieval Model=DPR2026.053946—
BiRDLLM Backbone=Llama-3.1-8B, Retrieval Model=ANCE2026.053964—
BiRDLLM Backbone=Mistral-7B, Retrieval Model=DPR2026.053848—
InstructRAGLLM Backbone=Llama-3.1-8B, Retrieval Model=DPR2026.053835—
BiRDLLM Backbone=Llama-3.1-8B, Retrieval Model=Contriever2026.053772—
ReliabilityRAGLLM Backbone=Qwen2.5-7B, Retrieval Model=Contriever2026.053639—
BiRDLLM Backbone=Qwen2.5-7B, Retrieval Model=Contriever2026.053660—
VanillaRAGLLM Backbone=Llama-3.1-8B, Retrieval Model=ANCE2026.053560—
VanillaRAGLLM Backbone=Qwen2.5-7B, Retrieval Model=ANCE2026.053449—
BiRDLLM Backbone=Qwen2.5-7B, Retrieval Model=ANCE2026.053449—
VanillaRAGLLM Backbone=Llama-3.1-8B, Retrieval Model=Contriever2026.053369—
DecodingLLM=Llama 2-C2025.0633—23
VanillaRAGLLM Backbone=Qwen2.5-7B, Retrieval Model=Contriever2026.053254—
DecodingLLM=Mistral-7B2025.0628—38
VanillaRAGLLM Backbone=Mistral-7B, Retrieval Model=ANCE2026.052749—
BiRDLLM Backbone=Mistral-7B, Retrieval Model=ANCE2026.052746—
VanillaRAGLLM Backbone=Mistral-7B, Retrieval Model=Contriever2026.052767—
BiRDLLM Backbone=Mistral-7B, Retrieval Model=Contriever2026.052759—
KeywordLLM=Llama 2-C2025.0625—36
RobustRAGLLM Backbone=Mistral-7B, Retrieval Model=Contriever2026.05226—
KeywordLLM=Mistral-7B2025.0621—48
RobustRAGLLM Backbone=Mistral-7B, Retrieval Model=ANCE2026.05194—
ReliabilityRAGLLM Backbone=Mistral-7B, Retrieval Model=ANCE2026.05194—
ReliabilityRAGLLM Backbone=Mistral-7B, Retrieval Model=Contriever2026.05184—
AV FilterLLM=GPT-4o, alpha=∞2025.0617.8—35.6
KeywordLLM=GPT-4o2025.0617.4—43.6
RobustRAGLLM Backbone=Mistral-7B, Retrieval Model=DPR2026.05175—
AV FilterLLM=Mistral-7B, alpha=∞2025.0613.4—47.2
ReliabilityRAGLLM Backbone=Mistral-7B, Retrieval Model=DPR2026.05134—
AV FilterLLM=Llama 2-C, alpha=52025.0611.4—34
AV FilterLLM=GPT-4o, alpha=102025.0611.2—40
AV FilterLLM=Llama 2-C, alpha=102025.0610.4—34.4
AV FilterLLM=GPT-4o, alpha=52025.069.8—42.6
AV FilterLLM=Mistral-7B, alpha=102025.068.4—52.6
AV FilterLLM=Mistral-7B, alpha=52025.068—53