Add pipeline tag and improve model card discovery (#1)

- Add pipeline tag and improve model card discovery (087f70d8d9666596ea992b77b4802cd578f1cbff)


Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>

README.md

@@ -1,87 +1,28 @@
 ---
 license: mit
+pipeline_tag: graph-ml
 ---
 
 <h1 align="center" style="font-size: 24px;">EquiformerV3:<br>Scaling Efficient, Expressive, and General SE(3)-Equivariant Graph Attention Transformers</h1> 
 
-<!--
-# **[Code](https://github.com/atomicarchitects/equiformer_v3)** | **[Paper]()**
--->
-
+<p align="center">
 <a href="https://github.com/atomicarchitects/equiformer_v3" style="color: #1a73e8; font-weight: bold; font-size: 20px;">Code</a> | 
 <a href="https://arxiv.org/abs/2604.09130" style="color: #1a73e8; font-weight: bold; font-size: 20px;">Paper</a>
-
-This repository contains the checkpoints of the work "EquiformerV3: Scaling Efficient, Expressive, and General SE(3)-Equivariant Graph Attention Transformers".
-Please refer to the [code](https://github.com/atomicarchitects/equiformer_v3) for detailed description of usage.
-
-<p align="center">
-  <img width="50%" height="50%" src="https://cdn-uploads.huggingface.co/production/uploads/64948a4a8d5ff0dd776655fe/03TPndezDyUw4FcfTBk4n.png"?
 </p>
 
+This repository contains the checkpoints for **EquiformerV3**, the third generation of the $SE(3)$-equivariant graph attention Transformer. EquiformerV3 is designed to advance efficiency, expressivity, and generality in 3D atomistic modeling.
 
+Building on EquiformerV2, this version introduces software optimizations achieving a $1.75\times$ speedup, structural improvements like equivariant merged layer normalization and smooth-cutoff attention, and SwiGLU-$S^2$ activations to incorporate many-body interactions while preserving strict equivariance. EquiformerV3 achieves state-of-the-art results on benchmarks including OC20, OMat24, and Matbench Discovery.
 
+Please refer to the [official GitHub repository](https://github.com/atomicarchitects/equiformer_v3) for detailed instructions on environment setup and usage.
 
-## Content ##
-<!-- 
-0. [OC20](#oc20)
--->
-0. [MPtrj](#mptrj)
-0. [OMat24 → MPtrj and sAlex](#oam)
-
-
-<!--
-<h2 id="oc20">OC20</h2>
-
-<table>
-  <tr style="background-color: #f0f0f0;">
-   <td><strong>Model</strong></td>
-   <td><strong>Training data</strong></td>
-   <td><strong>Config</strong></td>
-   <td><strong>Checkpoint</strong></td>
-  </tr>
-
-  <tr>
-   <td>EquiformerV3 (91M)</td>
-   <td>OC20 S2EF-2M</td>
-   <td><a href="https://github.com/atomicarchitects/equiformer_v3/blob/main/experimental/configs/oc20/2M/equiformer_v3/experiments/base_N%408-L%406-C%40128-attn-hidden%4064-ffn%40512-envelope-num-rbf%40128_merge-layer-norm_gates2-gridmlp_use-gate-force-head_wd%401e-3-grad-clip%40100_lin-ref-e%404.yml">base.yml</a></td>
-   <td></td>
-  </tr>
-</table>
--->
+<p align="center">
+  <img width="50%" height="50%" src="https://cdn-uploads.huggingface.co/production/uploads/64948a4a8d5ff0dd776655fe/03TPndezDyUw4FcfTBk4n.png">
+</p>
 
+## Checkpoints
 
-<h2 id="mptrj">MPtrj</h2>
-<!--
-Training consists of (1) direct pre-training and (2) gradient fine-tuning initialized from (1).
-<table>
-  <tr style="background-color: #f0f0f0;">
-   <td><strong>Model</strong></td>
-   <td><strong>Training data</strong></td>
-   <td><strong>Config</strong></td>
-   <td><strong>Checkpoint</strong></td>
-  </tr>
-  <tr>
-   <td>EquiformerV3 (direct pre-training)</td>
-   <td>MPtrj</td>
-   <td>
-     <a href="https://github.com/atomicarchitects/equiformer_v3/blob/main/experimental/configs/omat24/mptrj/experiments/direct/equiformer_v3_N%407_L%404_attn-hidden%4032_rbf%4010_max-neighbors%40300_attn-grid%4014-8_ffn-grid%4014_use-gate-force-head_merge-layer-norm_epochs%4070-bs%40512-wd%401e-3-beta2%400.95_dens-p%400.5-std%400.025-r%400.5-w%4010-strict-max-r%400.75-no-stress.yml">
-       direct.yml
-     </a>
-   </td>
-   <td></td>
-  </tr>
-  <tr>
-   <td>EquiformerV3 (gradient fine-tuning)</td>
-   <td>MPtrj</td>
-   <td>
-     <a href="https://github.com/atomicarchitects/equiformer_v3/blob/main/experimental/configs/omat24/mptrj/experiments/gradient/equiformer_v3_grad-finetune_N%407_L%404_attn-hidden%4032_rbf%4010_max-neighbors%40300_attn-grid%4014-8_ffn-grid%4014_pt-reg-dens-no-stress-strict-max-r%400.75-ft-no-reg_lr%400-5e-5-epochs%4010-bs%4064x8-wd%401e-3-beta2%400.95.yml">
-       gradient.yml
-     </a>
-   </td>
-   <td></td>
-  </tr>
-</table>
--->
+### MPtrj
 <table>
   <tr style="background-color: #f0f0f0;">
    <td><strong>Model</strong></td>
@@ -99,11 +40,9 @@ Training consists of (1) direct pre-training and (2) gradient fine-tuning initia
   </tr>
 </table>
 
+### OMat24 → MPtrj and sAlex
+Training consists of (1) direct pre-training on OMat24, (2) gradient fine-tuning on OMat24 initialized from (1), and (3) gradient fine-tuning on MPtrj and sAlex initialized from (2).
 
-<h2 id="oam">OMat24 → MPtrj and sAlex</h2>
-Training consists of (1) direct pre-training on OMat24, 
-(2) gradient fine-tuning on OMat24 initialized from (1), and
-(3) gradient fine-tuning on MPtrj and sAlex initialized from (2).
 <table>
   <tr style="background-color: #f0f0f0;">
    <td><strong>Model</strong></td>
@@ -153,4 +92,17 @@ Training consists of (1) direct pre-training on OMat24,
      </a>
    </td>
   </tr>
-</table>
+</table>
+
+## Citation
+
+If you find this work helpful, please consider citing:
+
+```bibtex
+@article{equiformer_v3,
+    title={EquiformerV3: Scaling Efficient, Expressive, and General SE(3)-Equivariant Graph Attention Transformers}, 
+    author={Yi-Lun Liao and Alexander J. Hoffman and Sabrina C. Shen and Alexandre Duval and Sam Walton Norwood and Tess Smidt},
+    journal={arXiv preprint arXiv:2604.09130},
+    year={2026}
+}
+```