List your company

Long-context Question Answering on MuSiQue (F1)

51.02F1 ScoreLongMab
7.579218.857130.13541.4129Aug 19, 2025Sep 17, 2025Oct 16, 2025Nov 14, 2025Dec 13, 2025Jan 11, 2026Feb 9, 2026

Evaluation results

MethodLinks
LongMabBackbone=Qwen3-8B2025.0851.0248
LongFaith-POBackbone=Qwen3-8B2025.0846.3541
LongMab-SFTBackbone=Qwen3-8B2025.0845.6241.5
Logo-POBackbone=Qwen3-8B2025.0844.9841
Vanilla LLMBackbone=Qwen3-8B2025.0844.2441.5
SeaLong-POBackbone=Qwen3-8B2025.0844.1142.5
LongReward-POBackbone=Qwen3-8B2025.0841.9440.5
LongAlpacaBackbone=Qwen3-8B2025.0838.0636.5
GLM-4.1V-9B-Thinking VERARAG Strategy=Attention-Guided RAG2026.0230.58—
GLM-4.1V-9B-Thinking ColPali RAGRAG Strategy=ColPali RAG2026.0230.24—
LongAlignBackbone=Qwen3-8B2025.0828.7629
GLM-4.1V-9B-Thinking Random RAGRAG Strategy=Random RAG2026.0228.71—
GLM-4.1V-9B-Thinking Embedding RAGRAG Strategy=Embedding RAG2026.0228.08—
GLM-4.1V-9B-ThinkingRAG Strategy=Direct (None)2026.0227.85—
GlyphRAG Strategy=Direct (None)2026.0224.87—
Qwen3-VL-8B-Instruct VERARAG Strategy=Attention-Guided RAG2026.0217.84—
Qwen3-VL-8B-Instruct OCR RAGRAG Strategy=OCR RAG2026.0214.7—
Qwen3-VL-8B-Instruct Random RAGRAG Strategy=Random RAG2026.0214.68—
Qwen3-VL-8B-InstructRAG Strategy=Direct (None)2026.029.25—