Update README.md

README.md

@@ -1,3 +1,97 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+datasets:
+- Zigeng/DMax-LLaDA-2.0-Mini-Code-Trajectories
+base_model:
+- inclusionAI/LLaDA2.0-mini
+---
+
+<div align="center">
+<h1>🚀 DMax: Aggressive Parallel Decoding for dLLMs</h1>
+  <div align="center">
+  <a href="https://github.com/czg1225/DMax/blob/main/LICENSE">
+    <img alt="Apache" src="https://img.shields.io/badge/License-Apache-4E94CE.svg">
+  </a>
+  <a href="https://github.com/czg1225/DMax">
+    <img src="https://img.shields.io/badge/Paper-Arxiv-darkred.svg" alt="Paper">
+  </a>
+  <a href="https://github.com/czg1225/DMax">
+    <img src="https://img.shields.io/badge/GitHub-Code-blue.svg?logo=github&" alt="GitHub">
+  </a>
+</div>
+</div>
+
+> **DMax: Aggressive Parallel Decoding for dLLMs**  
+> [Zigeng Chen](https://czg1225.github.io/chenzigeng99/), [Gongfan Fang](https://fangggf.github.io/), [Xinyin Ma](https://horseee.github.io/), [Ruonan Yu](https://scholar.google.com/citations?user=UHP95egAAAAJ&hl=en), [Xinchao Wang](https://sites.google.com/site/sitexinchaowang/)  
+> [xML Lab](https://sites.google.com/view/xml-nus), National University of Singapore
+
+
+## 💪 Highlights
+
+- **Aggressive Decoding Parallelism**: Achieves 6.0 TPF on math and reasoning tasks and 6.6 TPF on code tasks while preserving accuracy.
+- **Self-Revising dLLM**: Extends a pretrained MDLM into a UDLM with an intrinsic ability to revise its own erroneous predictions during decoding.
+- **Soft Parallel Decoding**: Uses interpolation between mask and token embeddings to propagate confidence priors from previous steps.
+
+<div align="center">
+  <img src="assets/tradeoff.png" width="90%" />
+  <br>
+  <em>Superior Parallelism-Accuracy Trade-off, Increased TPF with Maintained Accuracy.</em>
+</div>
+
+
+## 💻 Model and Datasets
+
+| Model | Description | Source Model | Link |
+| --- | --- | --- | --- |
+| 🤖 DMax-Math-16B | Highly parallel dLLM for math and reasoning. | LLaDA-2.0-mini | [Hugging Face](https://huggingface.co/Zigeng/DMax-Math-16B) |
+| 🤖 DMax-Coder-16B | Highly parallel dLLM for code generation. | LLaDA-2.0-mini | [Hugging Face](https://huggingface.co/Zigeng/DMax-Coder-16B) |
+
+| Dataset | Description | Link |
+| --- | --- | --- |
+| 📊 DMax-Math-Training-Data | Trajectories on math problems generated by LLaDA-2.0-mini | [Hugging Face](https://huggingface.co/datasets/Zigeng/DMax-LLaDA-2.0-Mini-Math-Trajectories) |
+| 📊 DMax-Code-Training-Data | Trajectories on code problems generated by LLaDA-2.0-mini | [Hugging Face](https://huggingface.co/datasets/Zigeng/DMax-LLaDA-2.0-Mini-Code-Trajectories) |
+
+
+## 🚀 Quick Start
+
+```python
+import torch
+from transformers import AutoModelForCausalLM
+from transformers import AutoTokenizer
+
+model = AutoModelForCausalLM.from_pretrained(
+    "Zigeng/DMax-Coder-16B", trust_remote_code=True, device_map="cuda:0"
+)
+model = model.to(torch.bfloat16)
+model.eval()
+tokenizer = AutoTokenizer.from_pretrained("Zigeng/DMax-Coder-16B", trust_remote_code=True)
+
+prompt = "Write a python function to find the first repeated character in a given string." + "\n\nPlease enclose your code within delimiters as follows:\n```python\n# YOUR CODE HERE\n```\n\n"
+
+input_ids = tokenizer.apply_chat_template(
+    [{"role": "user", "content": prompt}],
+    add_generation_prompt=True,
+    tokenize=True,
+    return_tensors="pt",
+)
+
+nfe, generated_tokens = model.generate_spd(
+    inputs=input_ids,
+    gen_length=2048,
+    block_length=32,
+    threshold=0.65,
+)
+
+generated_answer = tokenizer.decode(
+    generated_tokens[0],
+    skip_special_tokens=True,
+)
+
+print(generated_answer)
+print("nfe:",nfe,"token length",len(generated_tokens[0]))
+```
+
+## 📖 Experimental Results
+
+![trade-off](assets/exp.png)
+