List your company

Question Answering on 2Wiki Noisy

24.17F1 ScoreOpenDecoder
4.54529.640114.73519.8299Jan 13, 2026

Evaluation results

MethodLinks
OpenDecoderBackbone=Qwen-2.5-3B-Instruct2026.0124.1722.13
RbFTBackbone=Qwen-2.5-3B-Instruct2026.0123.622
Vanilla SFTBackbone=Qwen-2.5-3B-Instruct2026.0122.0720.4
AstuteRAGBackbone=Qwen-2.5-3B-Instruct2026.0120.8418.6
InstructRAGBackbone=Qwen-2.5-3B-Instruct2026.0115.39
Vanilla RAGBackbone=Qwen-2.5-3B-Instruct2026.0111.9720.38
RobustRAGBackbone=Qwen-2.5-3B-Instruct2026.015.38.2