Upload README.md with huggingface_hub

README.md

@@ -15,7 +15,7 @@ base_model: unsloth/Llama-3.1-8B-Instruct
 
 # EDT-Former: Full Model (Stage 2)
 
-The full **EDT-Former** model (DQ-Former encoder + Llama-3.1-8B-Instruct), as described in the ICLR 2026 paper:
+The full **EDT-Former** model (encoder + Llama-3.1-8B-Instruct), as described in the ICLR 2026 paper:
 
 > **Entropy-Guided Dynamic Tokens for Graph-LLM Alignment in Molecular Understanding**
 > Zihao Jing, Qiuhao Zeng, Ruiyi Fang, Yan Sun, Boyu Wang, Pingzhao Hu
@@ -23,7 +23,7 @@ The full **EDT-Former** model (DQ-Former encoder + Llama-3.1-8B-Instruct), as de
 
 ## Model Description
 
-EDT-Former aligns molecular graphs with a frozen LLM backbone (Llama-3.1-8B-Instruct) via a DQ-Former connector. Key properties:
+EDT-Former aligns molecular graphs with a frozen LLM backbone (Llama-3.1-8B-Instruct) via an entropy-guided dynamic token connector. Key properties:
 
 - **No LLM backbone fine-tuning** (only the embedding layer and connector are trained) — computationally efficient
 - **Entropy-guided dynamic token selection** preserves both local (substructural) and global molecular features
@@ -45,9 +45,9 @@ cp env.sh local.env.sh
 # Edit local.env.sh: set BASE_DIR, DATA_DIR, CHECKPOINT_DIR
 source local.env.sh
 
-# 3. Download model
+# 3. Download the model
 from huggingface_hub import snapshot_download
-snapshot_download("zihaojing/DQFormer-model", local_dir="checkpoints/edt_former_s2_large/final_model")
+snapshot_download("zihaojing/EDT-Former-model", local_dir="checkpoints/edt_former_s2_large/final_model")
 
 # 4. Run inference (example: forward reaction prediction)
 bash scripts/qa/mol_forward.sh
@@ -70,8 +70,8 @@ All evaluation scripts are in `scripts/qa/`. Example tasks:
 | Setting | Value |
 |---------|-------|
 | LLM backbone | Llama-3.1-8B-Instruct (frozen) |
-| Stage 1 encoder | [zihaojing/DQFormer-encoder](https://huggingface.co/zihaojing/DQFormer-encoder) |
-| Training data | [zihaojing/DQFormer-sft-data](https://huggingface.co/datasets/zihaojing/DQFormer-sft-data) |
+| Stage 1 encoder | [zihaojing/EDT-Former-encoder](https://huggingface.co/zihaojing/EDT-Former-encoder) |
+| Training data | [zihaojing/EDT-Former-sft-data](https://huggingface.co/datasets/zihaojing/EDT-Former-sft-data) |
 | Epochs | 2 |
 | Learning rate | 1e-4 (cosine) |
 | Batch size | 4 × 8 grad accum = effective 32 |
@@ -81,9 +81,9 @@ All evaluation scripts are in `scripts/qa/`. Example tasks:
 
 | Resource | Link |
 |----------|------|
-| Pretrain Data | [zihaojing/DQFormer-pretrain-data](https://huggingface.co/datasets/zihaojing/DQFormer-pretrain-data) |
-| SFT Data | [zihaojing/DQFormer-sft-data](https://huggingface.co/datasets/zihaojing/DQFormer-sft-data) |
-| Encoder (Stage 1) | [zihaojing/DQFormer-encoder](https://huggingface.co/zihaojing/DQFormer-encoder) |
+| Pretrain Data | [zihaojing/EDT-Former-pretrain-data](https://huggingface.co/datasets/zihaojing/EDT-Former-pretrain-data) |
+| SFT Data | [zihaojing/EDT-Former-sft-data](https://huggingface.co/datasets/zihaojing/EDT-Former-sft-data) |
+| Encoder (Stage 1) | [zihaojing/EDT-Former-encoder](https://huggingface.co/zihaojing/EDT-Former-encoder) |
 | Code | [selmiss/DQ-Former](https://github.com/selmiss/DQ-Former) |
 
 ## Citation