List your company

Document Question Answering on MMLongBench-Doc

65.8AccuracyHuman Baseline
27.52837.46447.457.336Jan 8, 2026Feb 5, 2026Mar 6, 2026Apr 3, 2026May 2, 2026May 30, 2026Jun 28, 2026

Evaluation results

MethodLinks
Human Baseline2026.0165.866————
Evolved AgentModel=Gemini-3.1-Flash, Evolved=true2026.0662—————
MARDocBackbone=Qwen3-VL-32B2026.0659.957.5————
Evolved Agent, No Ground TruthModel=Gemini-3.1-Flash, Evolved=true2026.0658.4—————
Evolved Agent, No ExplorationModel=Gemini-3.1-Flash, Evolved=true2026.0657.4—————
DocAgentModel=Claude-3.5-Sonnet2026.0157.354.1————
MARDocBackbone=Qwen3-VL-30B-A3B2026.0657.154.6————
DocDancerModel=GPT-5.22026.015756.867.6———
SimpleDocModel=Gemini-3.1-Flash, Evolved=false2026.0656.6—————
DocDancerModel=Gemini-2.5-Pro2026.0156.355.365.9———
Evolved Agent, GPT-5-minimal Meta AgentModel=Gemini-3.1-Flash, Evolved=true2026.0656.1—————
Evolved Agent, No ExplorationModel=GPT-5-mini, Evolved=true2026.0655.5—————
Evolved AgentModel=GPT-5-mini, Evolved=true2026.0655.4—————
Evolved Agent, No Ground TruthModel=GPT-5-mini, Evolved=true2026.0654.9—————
DocDancerModel=Qwen3-30B-A3B (ft)2026.0154.453.965.3———
MACTModel=Gemini-3.1-Flash, Evolved=false2026.0654.4—————
Evolved Agent, GPT-5-minimal Meta AgentModel=GPT-5-mini, Evolved=true2026.0652.8—————
DocDancerModel=GPT-4o2026.0152.350.859.2———
DocAgentModel=GPT-4o2026.0151.849.1————
SimpleDocModel=GPT-5-mini, Evolved=false2026.0651.3—————
MDocAgentModel=Gemini-3.1-Flash, Evolved=false2026.0651.2—————
DocDancerModel=Qwen3-4B (ft)2026.0148.449.259.4———
MACTModel=MiMo-VL-7B2026.0147.4—————
MACTModel=GPT-5-mini, Evolved=false2026.0647.4—————
File APIModel=Gemini-3.1-Flash, Evolved=false2026.0645.4—————
File APIModel=GPT-5-mini, Evolved=false2026.0644.6—————
MDocAgentModel=GPT-5-mini, Evolved=false2026.0644—————
Naive VLModel=GPT-4o2026.0142.844.9————
RAGAnythingModel=GPT-4o-mini2026.0142.8—————
Baseline AgentModel=Gemini-3.1-Flash, Evolved=false2026.0642.4—————
MDocAgentModel=GPT-4o2026.0142—————
Baseline AgentModel=GPT-5-mini, Evolved=false2026.0640.7—————
CoTModel=Gemini-3.1-Flash, Evolved=false2026.0640.4—————
TesseractModel=Gemini-2.0-Flash2026.0139.637.2————
Doc-ReactModel=GPT-4o2026.0138.138.3————
CoTModel=GPT-5-mini, Evolved=false2026.0636.2—————
ColpaliModel=GPT-4o2026.0132.230.8————
M3DocRAG w/ ColPaliModel=Qwen2-VL-7B2026.0131.436.5————
TesseractModel=GPT-4o2026.0130.130.5————
VisRAGModel=GPT-4o2026.012927.8————
DocLensModel=Claude-4-Sonnet2026.01——63.3———
DocLensModel=Gemini-2.5-Pro2026.01——67.6———
GPT-4o-200b-128Retrieval=Ground-Truth2025.10———69.669.356.5
GPT-4o-200b-128Retrieval=retrieving@52025.10———57.560.730.3
GPT-4o-200b-128Retrieval=retrieving@102025.10———6163.737.2
GPT-4o-200b-128Retrieval=topk-adjusted2025.10———59.362.940.9
GPT-4o-200b-128Retrieval=LAD-RAG2025.10———62.567.645
Idefics2-8BRetrieval=All-Pages2025.10———77.77.2
InternVL2-8BRetrieval=Ground-Truth2025.10———39.950.625
InternVL2-8BRetrieval=retrieving@52025.10———28.737.216.4
InternVL2-8BRetrieval=retrieving@102025.10———31.939.520.8
InternVL2-8BRetrieval=topk-adjusted2025.10———30.436.521.2
InternVL2-8BRetrieval=LAD-RAG2025.10———44.849.524.2
MiniCPM-Llama3-V2.5-8BRetrieval=All-Pages2025.10———8.59.59.5
mPLUG-DocOwl 1.5-8BRetrieval=All-Pages2025.10———6.97.46.4
Naive VLModel=Gemini-2.5-Pro2026.01——58.1———
Phi-3.5-Vision-4BRetrieval=Ground-Truth2025.10———38.349.222.7
Phi-3.5-Vision-4BRetrieval=retrieving@52025.10———31.539.919.2
Phi-3.5-Vision-4BRetrieval=retrieving@102025.10———30.238.118.9
Phi-3.5-Vision-4BRetrieval=topk-adjusted2025.10———33.64019.9
Phi-3.5-Vision-4BRetrieval=LAD-RAG2025.10———39.141.420.2
Pixtral-12B-2409Retrieval=Ground-Truth2025.10———49.853.734.3
Pixtral-12B-2409Retrieval=retrieving@52025.10———38.939.520
Pixtral-12B-2409Retrieval=retrieving@102025.10———34.532.717.8
Pixtral-12B-2409Retrieval=topk-adjusted2025.10———38.339.421.3
Pixtral-12B-2409Retrieval=LAD-RAG2025.10———41.546.226.1
SimpleDocModel=Claude-4-Sonnet2026.01——58.6———
SimpleDocModel=Gemini-2.5-Pro2026.01——56.6———