rank,id,method,kind,retrieval_top1,retrieval_top5,cosine_matched,knn_mae_heat_all,knn_mae_dir_gap,l2_matched,cosine_encoder,cosine_projection,n_profiles,n_evaluated,parameters,training_data,date,paper,code,weights
1,meidnet-2k,MEIDNet (published model),model,0.33606340819022457,0.8964332892998679,0.7717390060424805,0.021445706737120214,0.05529194187582562,0.6753583550453186,-0.008391765877604485,0.7717390060424805,367.0,3785.0,702405,"Perov-5, all 18,928 materials",2026-10-03,https://doi.org/10.1038/s41524-026-02153-3,https://github.com/ABnano/MEIDNet,https://huggingface.co/Babu09/MEIDNet/blob/main/dual_autoencoder_clip_earlyfusion_propertyaware_2k.pth
2,meidnet-alignment,"MEIDNet (alignment training, 7 seeds)",model,0.2972636346480468,0.8551047367427816,0.9535511391503471,0.02514406491790904,0.04922664653708247,0.3015082861695971,0.9535511391503471,-0.9782245329448155,367.0,3785.0,702405,"Perov-5, all 18,928 materials",2026-10-05,https://doi.org/10.1038/s41524-026-02153-3,https://github.com/ABnano/MEIDNet,https://huggingface.co/Babu09/MEIDNet/tree/main/reproduction
3,meidnet-propertyaware,MEIDNet (shorter training),model,0.29669749009247026,0.8480845442536328,0.48043641448020935,0.024196565389696178,0.046900924702774106,1.019321322441101,0.002863099565729499,0.48043641448020935,367.0,3785.0,702405,"Perov-5, all 18,928 materials",2026-10-03,https://doi.org/10.1038/s41524-026-02153-3,https://github.com/ABnano/MEIDNet,https://huggingface.co/Babu09/MEIDNet/blob/main/dual_autoencoder_clip_earlyfusion_propertyaware.pth
4,meidnet-earlyfusion,MEIDNet (first early-fusion model),model,0.02774108322324967,0.13130779392338177,0.6550266742706299,0.05149062087186262,0.06434874504623514,0.8223259449005127,0.6550266742706299,-0.9787194728851318,367.0,3785.0,702405,"Perov-5, all 18,928 materials",2026-10-03,https://doi.org/10.1038/s41524-026-02153-3,https://github.com/ABnano/MEIDNet,https://huggingface.co/Babu09/MEIDNet/blob/main/dual_autoencoder_clip_earlyfusion.pth
5,baseline-chance,Chance level,baseline,0.0027247956403269754,0.013623978201634877,0.0,,,,,,367.0,3785.0,,,2026-10-03,,https://github.com/ABnano/MEIDNet,
6,baseline-composition-knn,Composition k-NN,baseline,,,,0.5106261558784676,0.16282166446499338,,,,,3785.0,,"Perov-5 train (11,356)",2026-10-03,,https://github.com/ABnano/MEIDNet,