EquiformerV3+DeNS-OAM

Version: v2026.04.07 Added: 2026-04-07 Published: 2026-04-07 30.3M parameters Missing preds: 0
Repo Paper Docs DOI Files PR Checkpoint
Predictions
κSRMEGeometry OptimizationDiscovery

ML vs DFT Formation Energies

ML vs DFT Convex Hull Distance

ML vs DFT Lattice Thermal Conductivity

Convex hull distance prediction errors projected onto elements

1 H 0.11
2 He 0.00
3 Li 0.02
4 Be 0.01
5 B 0.04
6 C 0.04
7 N 0.06
8 O 0.10
9 F 0.09
10 Ne 0.00
11 Na 0.02
12 Mg 0.03
13 Al 0.03
14 Si 0.04
15 P 0.04
16 S 0.05
17 Cl 0.07
18 Ar 0.00
19 K 0.03
20 Ca 0.03
21 Sc 0.02
22 Ti 0.03
23 V 0.05
24 Cr 0.07
25 Mn 0.10
26 Fe 0.08
27 Co 0.04
28 Ni 0.03
29 Cu 0.03
30 Zn 0.03
31 Ga 0.03
32 Ge 0.04
33 As 0.04
34 Se 0.07
35 Br 0.06
36 Kr 0.00
37 Rb 0.03
38 Sr 0.03
39 Y 0.03
40 Zr 0.03
41 Nb 0.03
42 Mo 0.04
43 Tc 0.03
44 Ru 0.04
45 Rh 0.04
46 Pd 0.04
47 Ag 0.03
48 Cd 0.02
49 In 0.04
50 Sn 0.03
51 Sb 0.04
52 Te 0.10
53 I 0.04
54 Xe 0.00
55 Cs 0.03
56 Ba 0.03
57 La 0.03
58 Ce 0.02
59 Pr 0.03
60 Nd 0.02
61 Pm 0.03
62 Sm 0.02
63 Eu 0.06
64 Gd 0.04
65 Tb 0.02
66 Dy 0.03
67 Ho 0.02
68 Er 0.02
69 Tm 0.02
70 Yb 0.03
71 Lu 0.02
72 Hf 0.02
73 Ta 0.06
74 W 0.04
75 Re 0.03
76 Os 0.04
77 Ir 0.05
78 Pt 0.05
79 Au 0.05
80 Hg 0.02
81 Tl 0.03
82 Pb 0.05
83 Bi 0.03
84 Po 0.00
85 At 0.00
86 Rn 0.00
87 Fr 0.00
88 Ra 0.00
89 Ac 0.03
90 Th 0.03
91 Pa 0.03
92 U 0.04
93 Np 0.08
94 Pu 0.19
95 Am 0.00
96 Cm 0.00
97 Bk 0.00
98 Cf 0.00
99 Es 0.00
100 Fm 0.00
101 Md 0.00
102 No 0.00
103 Lr 0.00
104 Rf 0.00
105 Db 0.00
106 Sg 0.00
107 Bh 0.00
108 Hs 0.00
109 Mt 0.00
110 Ds 0.00
111 Rg 0.00
112 Cn 0.00
113 Nh 0.00
114 Fl 0.00
115 Mc 0.00
116 Lv 0.00
117 Ts 0.00
118 Og 0.00
57-71 La-Lu Lanthanides
89-103 Ac-Lr Actinides

Model Authors

  1. Yi-Lun Liao  Massachusetts Institute of Technology  
  2. Alexander J. Hoffman  Mirror Physics  
  3. Sabrina C. Shen  Mirror Physics  
  4. Sam Walton Norwood  Mirror Physics  
  5. Tess Smidt  Massachusetts Institute of Technology  

Trained By

  1. Yi-Lun Liao  Massachusetts Institute of Technology  

Model Info

  • Model Version v2026.04.07
  • Model Type UIP
  • Targets EFSG
  • Openness OSOD
  • Train Task S2EFS
  • Test Task IS2RE-SR
  • Trained for Benchmark Yes

Training Set

OMat24: 101M structures from 3.23M materials

MPtrj: 1.58M structures from 146k materials

Subsampled Alexandria: 10.4M structures from 3.23M materials

Description

EquiformerV3 is the third generation of SE(3)-equivariant graph attention Transformers. Please refer to our paper for more details.

Hyperparameters

  • max_force: 0.02
  • max_steps: 500
  • ase_optimizer: "FIRE"
  • cell_filter: "FrechetCellFilter"
  • graph_construction_radius: 6
  • max_neighbors: null
  • num_blocks: 7
  • lmax: 4
  • mmax: 2
  • embedding_dimension: 128
  • attention_value_hidden_size: 32
  • num_attention_heads: 8
  • attention_alpha_hidden_size: 64
  • attention_value_size_per_head: 16
  • ffn_hidden_size_after_swiglu_s2_activation: 512
  • grid_resolution_attention: "(14, 8)"
  • grid_resolution_ffn: "(14, 14)"
  • DeNS_probability: 0.5
  • DeNS_coefficient: 1
  • DeNS_noise_std: 0.025
  • DeNS_corruption_ratio: 0.5
  • DeNS_max_force_norm: 2.5

Dependencies