Update README.md

README.md

@@ -2,30 +2,21 @@
 
 ## 1. Overview
 
-RLM (Recipe Language Model) is a domain-specific language model designed for recipe learning, recommendation and mechanistic reasoning in materials research. With perovskite solar cells used in this work as a demonstration system.
+Here, we present a domain-specific recipe language model (RLM) developed for an emerging scientific tool of robotic boxes, with perovskite solar cell research as a demonstration. For iterative fine-tuning of the RLM, seven artificial intelligence (AI) layers, including learning, generating, RecipeQA, fine-tuning, reasoning, evaluation, and optimisation, have been designed with a language agent. During the loops of seven AI layers, both numerical and semantic recipes are continuously learned and optimised for the RLM.
 
 ---
 
 ## 2. Model Description
 
-RLM (Recipe Language Model) is trained to understand and generate structured and natural-language experimental procedures for scientific research.
-
-The model supports:
-
-- Experimental recipe generation
-- Process parameter optimization
-- Mechanism-aware reasoning
-
-It is coordinated by a language agent and iteratively improved through a seven-layer AI system.
+To train this domain-specific RLM, the workflow starts from encoded formulas and parameters as recipe inputs, proceeds through seven AI layers with the language agent, and produces in situ characterisation and device performance assessment as mechanistic outputs. The language agent encodes these machine-readable recipes into structured formulas and parameters sequences, which are translated into tokens for subsequent fine-tuning of the RLM and execution by the robotic boxes. As a result, the fine-tuned RLM incorporates the encoded recipes, robotics, and characterised results to form a closed recommendation–synthesis–fabrication–characterisation–mechanism loop for exploring the large space of the recipes and their underlying mechanisms.
 
 ---
 
 ## 3. Key Features
 
-- Domain-specific modeling for scientific experimentation
-- Recipe generation for experimental workflows
-- Mechanism-informed reasoning (physics + chemistry)
-- Closed-loop optimization with experimental feedback
+Recipe recommendation: the ability to generate recommended recipes with recipe integrity, formula rationality, parameter rationality, and experimental validity.
+
+Mechanistic reasoning: the ability to provide mechanistic explanations with domain knowledge, mechanism integrity, interpretation, comprehensiveness, and coherence.
 
 ---
 
@@ -35,29 +26,25 @@ The model is trained using a multi-stage pipeline:
 
 - Data sources:
   - Scientific literature
-  - Experimental datasets
-  - Generated RecipeQA data
+  - Robotic experimental datasets
+  - Generated RecipeQA 
 - Training strategy:
   - Supervised fine-tuning (SFT with LoRA)
-  - Preference optimization (DPO)
+  - Direct preference optimisation (DPO) 
 - Data format:
-  - Question-answer (RecipeQA)
-  - Mechanism reasoning data (Chain-of-Thought)
-
-The training process is organized within a seven-layer AI architecture:
-Learning → Generating → RecipeQA → Fine-tuning → Reasoning → Evaluation → Optimization
+  - RecipeQA without mechanism
+  - RecipeQA with mechanism (Chain-of-Thought) 
 
 ---
 
 ## 5. Intended Use
 
-This model is intended for:
+RLM is designed to assist researchers, scientists, and industry professionals in recipe-driven domains that involve the vast combinatorial exploration of recipes and are often constrained by time-consuming trial-and-error synthesis and labour-intensive fabrication. 
 
-- Materials science research
-- AI-assisted experimental design
-- Process optimization
-- Integration with laboratory automation systems
+Representative application scenarios include: 
+- Biomedicine 
+- Synthetic Biology
+- New-Energy Batteries
+- Semiconductor Materials
 
 ---
-
-