update readme

README.md

@@ -1,3 +1,33 @@
 ---
 license: apache-2.0
+tags:
+- chemistry
+- biology
 ---
+# ByteFF2
+
+This repository contains the model used for the paper [Bridging Quantum Mechanics to Organic Liquid Properties via a Universal Force Field](https://arxiv.org/abs/2508.08575)。
+
+[ByteFF-Pol](https://arxiv.org/abs/2508.08575) is a polarizable force field parameterized by a graph neural network (GNN), trained on high-level quantum mechanics (QM) data, thus eliminating the need for experimental calibration. ByteFF-Pol achieves exceptional accuracy in predicting the thermodynamic and transport properties of small-molecule liquids and electrolytes, outperforming SOTA traditional and ML force fields
+
+# Trained Models
+The `trained_models`` folder contains the trained model for ByteFF-Pol and its corresponding configuration (.yaml) file.
+
+# How to use
+Code and examples are available in the [byteff2](https://github.com/ByteDance-Seed/byteff2) repository.
+
+## Citation
+If you find ByteFF-Pol is useful for your research and applications, feel free to give us a star ⭐ or cite us using:
+
+```bibtex
+
+@misc{zheng2025bridgingquantummechanicsorganic,
+  title         = {Bridging Quantum Mechanics to Organic Liquid Properties via a Universal Force Field},
+  author        = {Tianze Zheng and Xingyuan Xu and Zhi Wang and Xu Han and Zhenliang Mu and Ziqing Zhang and Sheng Gong and Kuang Yu and Wen Yan},
+  year          = {2025},
+  eprint        = {2508.08575},
+  archivePrefix = {arXiv},
+  primaryClass  = {physics.comp-ph},
+  url           = {https://arxiv.org/abs/2508.08575}
+}
+```

trained_models/fftrainer_config_in_use.yaml

@@ -0,0 +1,67 @@
+model:
+  ff_block:
+  - act: gelu
+    out_mlp_dims:
+    - 256
+    - 4
+    post_mlp_dims:
+    - 256
+    - 256
+    - 1
+    pre_mlp_dims:
+    - 256
+    - 256
+    - 3
+    tanh_output: 20.0
+    type: MMBondedConj
+  - act: gelu
+    out_mlp_dims:
+    - 256
+    - 3
+    pre_mlp_dims:
+    - 256
+    - 256
+    - 3
+    type: ChargeVolume
+  - act: gelu
+    charge14: 0.5
+    disp_damping_factor: 120.0
+    fix_li_alpha: 1.0e-06
+    ind14: 0.5
+    li_damp_clip: 0.2
+    out_mlp_dims:
+    - 256
+    - 3
+    pol_damping_factor: 0.39
+    pre_mlp_dims:
+    - 256
+    - 256
+    - 3
+    s12: 1.5
+    type: Exp6Pol
+    vdw14: 0.5
+  graph_block:
+    feature_layer:
+      act: gelu
+      atom_embedding_dim: 32
+      edge_mlp_dims:
+      - 64
+      - 256
+      - 2
+      node_mlp_dims:
+      - 64
+      - 256
+      - 2
+    gnn_layer:
+      act: gelu
+      at_channels: 16
+      ffn_dims:
+      - 256
+      - 2
+      gnn_dims:
+      - 256
+      - 256
+      - 4
+      gnn_type: EGT
+      heads: 16
+      jk: cat

trained_models/optimal.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae47a6e6860b563908a2e0a83d4a3f6adc1c36b48f544e2241d24066d43d539c
+size 111892932