rank,id,method,kind,sun_rate,stable_rate,unique_rate,novel_rate,dhf_median,dft_hit_rate,dft_known,n_delivered,valid_rate,n_sun,n_budget,n_generated,parameters,training_data,date,paper,code,weights
1,baseline-screening,Encoder screening,baseline,0.8518518518518519,0.8518518518518519,1.0,1.0,-1.2835,,0.0,54.0,1.0,46.0,54.0,,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,baseline-random,Random sampling,baseline,0.7777777777777778,0.8888888888888888,1.0,0.8888888888888888,-1.3418999999999999,0.0,6.0,54.0,1.0,42.0,54.0,,,,2026-10-03,,https://github.com/ABnano/MEIDNet,
3,meidnet-earlyfusion,MEIDNet (first early-fusion model),model,0.6481481481481481,0.9444444444444444,1.0,0.7037037037037037,-1.287375,0.1875,16.0,54.0,1.0,35.0,54.0,,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
4,meidnet-2k,MEIDNet (published model),model,0.6111111111111112,0.9622641509433962,1.0,0.660377358490566,-1.51814,0.05555555555555555,18.0,53.0,0.9814814814814815,33.0,54.0,,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
5,meidnet-alignment-seed3,"MEIDNet (alignment training, seed 3)",model,0.6111111111111112,0.8723404255319149,1.0,0.8297872340425532,-1.31954,0.25,8.0,47.0,0.8703703703703703,33.0,54.0,,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/blob/main/reproduction/meidnet_paper_rerun_seed3.pth
6,meidnet-propertyaware,MEIDNet (shorter training),model,0.5555555555555556,0.9230769230769231,1.0,0.6538461538461539,-1.499765,0.05555555555555555,18.0,52.0,0.9629629629629629,30.0,54.0,,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