Update README.md

README.md

@@ -1,4 +1,52 @@
-SurfaceScienceAssistant: A Fine-Tuned Llama-3.1-8B for On-Surface Synthesis Reasoning1. Overview / 概述SurfaceScienceAssistant is a specialized Large Language Model (LLM) fine-tuned for high-fidelity reasoning and knowledge synthesis in the field of Surface Chemistry and On-Surface Synthesis (OSS).While general-purpose models often struggle with the niche nomenclature and complex mechanistic logic of surface-confined reactions, this model has been specifically adapted to understand the synergistic interactions between organic precursors and metal substrates (e.g., Au, Ag, Cu). It serves as the intelligent reasoning core for the OSS Assistant platform, providing expert-level insights into reaction pathways, molecular self-assembly, and scanning probe microscopy (SPM) interpretation.2. Model Specifications / 模型规格Developer: Materials Genome Institute (MGI), Shanghai University.Architecture: Meta-Llama-3.1-8B (Transformer-based).Adaptation Method: QLoRA (4-bit Low-Rank Adaptation) — ensuring a balance between maintaining foundational scientific priors and acquiring domain-specific expertise.Primary Objective: Domain-Specific Question Answering (QA) and Mechanistic Reasoning. (Note: This model is optimized for scientific dialogue rather than structured data extraction).Training Infrastructure: BitsAndBytes 4-bit NormalFloat (NF4) quantization with gradient checkpointing for resource-efficient, high-performance training.3. Training & Dataset / 训练与数据The model underwent instruction tuning using a curated Surface-Chemistry-Knowledge-Graph (SCKG) dataset, which includes:Reaction Mechanisms: Detailed reasoning chains for Ullmann coupling, Glaser coupling, and Graphene Nanoribbon (GNR) synthesis.Substrate-Specific Effects: Specialized knowledge on how different crystalline faces (e.g., (111), (110)) influence molecular diffusion and activation energy.Expert Dialogue: Synthetic and human-annotated QA pairs covering DFT calculation parameters (VASP/CP2K) and experimental troubleshooting.Hyperparameters:ParameterValueDescriptionLoRA Rank ($r$)16Capturing domain-specific manifoldsLoRA Alpha ($\alpha$)32Scaling factor for stable weight updatesLearning Rate2e-4Optimized for Llama-3.1 backboneEffective Batch Size4Achieved via Gradient Accumulation4. Key Capabilities / 核心能力Expert-Level QA: Provides grounded answers regarding surface-confined reaction conditions and catalyst-precursor interactions.Mechanistic Insight: Capable of reasoning through multi-step debromination or cyclodehydrogenation processes.Context-Aware Dialog: In conjunction with a History-Aware Retriever, the model maintains temporal consistency over complex, multi-turn scientific discussions.Scientific Trustworthiness: Designed to work within a RAG (Retrieval-Augmented Generation) framework to minimize hallucinations by grounding responses in primary literature.5. Usage & Implementation / 使用说明This model is best utilized with a system prompt that defines its role as a "Surface Science Expert."Example Inference (Python):Pythonfrom transformers import AutoModelForCausalLM, AutoTokenizer
+# SurfaceScienceAssistant: A Fine-Tuned Llama-3.1-8B for On-Surface Synthesis Reasoning
+
+[![Model License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
+[![Base Model](https://img.shields.io/badge/Base%20Model-Llama--3.1--8B-orange)](https://huggingface.co/meta-llama/Llama-3.1-8B)
+[![Field](https://img.shields.io/badge/Field-Surface%20Chemistry-green)](https://en.wikipedia.org/wiki/Surface_science)
+
+## 1. Overview
+**SurfaceScienceAssistant** is a specialized Large Language Model (LLM) fine-tuned for reasoning and knowledge synthesis in the field of **Surface Chemistry** and **On-Surface Synthesis (OSS)**. 
+
+While general-purpose models often struggle with the niche nomenclature and complex mechanistic logic of surface-confined reactions, this model has been specifically adapted to understand the synergistic interactions between organic precursors and metal substrates (e.g., Au, Ag, Cu). It serves as the intelligent reasoning core for the **OSS Assistant** platform, providing expert-level insights into reaction pathways.
+
+---
+
+## 2. Model Specifications
+* **Developer:** Materials Genome Institute (MGI), Shanghai University.
+* **Architecture:** Meta-Llama-3.1-8B (Transformer-based).
+* **Adaptation Method:** **QLoRA (4-bit Low-Rank Adaptation)** — ensuring a balance between maintaining foundational scientific priors and acquiring domain-specific expertise.
+* **Primary Objective:** **Targeted Scientific Reasoning** — This model has been specifically fine-tuned to master the intricate nuances of surface science, enabling dialogue regarding surface-confined chemical reactions, substrate-molecule interaction
+* **Training Infrastructure:** BitsAndBytes 4-bit NormalFloat (NF4) quantization with gradient checkpointing for resource-efficient, high-performance training.
+
+---
+## 3. Training & Dataset
+The model underwent domain-specific instruction tuning using the [**Surface_Chemistry**](https://huggingface.co/datasets/JuanXiang-SHU/Surface_Chemistry) dataset. This curated collection integrates high-quality instruction-following pairs derived from extensive surface science literature and expert-annotated reasoning chains.
+* **Mechanistic Reasoning:** On-surface reactions and molecular self-assembly, facilitating precise functionalization and structural evolution.
+* **Substrate Effects:** The influence of crystalline facets and electronic structures on molecular behavior.
+
+
+### Hyperparameters:
+| Parameter | Value | Description |
+| :--- | :--- | :--- |
+| **LoRA Rank ($r$)** | 16 | Capturing domain-specific manifolds |
+| **LoRA Alpha ($\alpha$)** | 32 | Scaling factor for stable weight updates |
+| **Learning Rate** | 2e-4 | Optimized for Llama-3.1 backbone |
+| **Effective Batch Size** | 4 | Achieved via Gradient Accumulation |
+
+---
+
+## 4. Key Capabilities
+**Expert-Level QA:** Provides grounded answers regarding surface-confined reaction conditions and catalyst-precursor interactions.
+**Mechanistic Insight:** Capable of reasoning through multi-step debromination or cyclodehydrogenation processes.
+
+---
+
+## 5. Usage & Implementation
+This model is best utilized with a system prompt that defines its role as a "Surface Science Expert."
+
+### Example Inference (Python):
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
 model_id = "JuanXiang-SHU/SurfaceScienceAssistant"
 tokenizer = AutoTokenizer.from_pretrained(model_id)
@@ -7,18 +55,4 @@ model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", load_i
 messages = [
     {"role": "system", "content": "You are a professional researcher in Surface Chemistry."},
     {"role": "user", "content": "How does the Au(111) substrate influence the Ullmann coupling of 1,4-dibromobenzene compared to Cu(111)?"}
-]
-
-# Standard inference pipeline follows...
-6. Limitations & Disclaimer / 局限性与免责声明Validation Required: As with all AI models, the chemical predictions and mechanistic proposals should be treated as hypotheses and verified via experimental methods or high-level DFT simulations.Domain Scope: While proficient in surface science, the model’s performance in unrelated fields (e.g., macro-economics, creative writing) may be diminished compared to the base Llama-3.1 model.7. Citation & Contact / 引用与联系If you find this model useful for your research, please cite:代码段@software{SurfaceScienceAssistant2026,
-  author = {Juan Xiang, Hao Jiang, and the Cosine Intelligence Team},
-  title = {SurfaceScienceAssistant: A Specialized LLM for On-Surface Synthesis Discovery},
-  year = {2026},
-  publisher = {Hugging Face},
-  url = {https://huggingface.co/JuanXiang-SHU/SurfaceScienceAssistant}
-}
-Contact: Juan Xiang | Materials Genome Institute, Shanghai University.
-
----
-license: mit
----
+]