Update README.md

README.md

@@ -47,104 +47,53 @@ language:
 
 # 1. Introduction
    
-We introduces Crab, a novel Configurable Role-Playing (RP) LLM with Assessing Benchmark, which consists of Role-Centric Dataset Curation, Persona-Embodying LLM Construction, and Comprehensive Benchmark Creation for RP dialogue generation.
-Distinct from traditional RP models that employ only several preset roles, Crab enables dynamic configuration of desired roles, thereby enhancing related flexibility and adaptability. 
-To effectively train RP-LLMs, we curated the largest RP training dataset. 
-The dataset provides a detailed role overview for each dialogue, including character profile, conversation scenario, and tagged topic, capturing a broad range of role-based behaviors, emotions, and interactions.
-We also noticed that current benchmarks lack both proper evaluation standards and methods. 
-Thus, to validate RP-LLMs' effectiveness, we introduced a new benchmark containing an evaluation standard, a test dataset with manual annotations, and a reward model RoleRM designed to automatically assess specific aspects of RP while aligning with human perception.
-Sufficient experiments reveal that RoleRM significantly outperforms ChatGPT and other evaluation methods in conducting fine-grained evaluations of RP. 
-Also, RP-LLMs powered by Crab demonstrate superior performance across various fine-grained aspects.
+This is a fine-tuned evaluator for roly-playing tasks. The used training data set can be seen in [Annotated Role-playing Evaluation Dataset](https://huggingface.co/datasets/HeAAAAA/Crab-manually-annotated-role-playing-evaluation-dataset). More details can be seen in [Crab](https://huggingface.co/HeAAAAA/Crab).
 
-More details can be seen at Github {https://github.com/KaiHe-better/Crab?tab=readme-ov-file}.
 
 
 
+# 2. Six Aspect-specific Metrics
 
-# 2. Configurable Role-Playing LLM
+- **Language Fluency**: Pertains to the natural and fluent communication style, independent of grammatical strictness or contextual background.
 
-<div align="center">
+- **Language Relevance**: Focuses on the ability to stay on topic and respond appropriately, essentially testing the capacity to follow instructions.
 
-  <!-- Logo 图片 -->
-  <img src="https://cdn-uploads.huggingface.co/production/uploads/650add6348983c90ab688b6e/fDaDq8tzBBUuEteND8N53.png" width="500" style="border-radius: 20px;"/>
+- **Role Language**: Evaluates whether the text reflects the vocabulary and tone specific to roles, including appropriate actions.
 
-</div>
+- **Role Knowledge**: Involves a deep understanding of both general knowledge and information specific to the roles, ensuring accurate and informed role portrayal.
 
+- **Emotional Expression**: Reviews the suitability of emotions, emotional intelligence, and empathy expressed in context with the role's traits.
 
-Unlike existing RP-LLMs, where a single role is trained with numerous dialogues, our approach introduces a diverse range of roles with detailed configuration information
-while keeping dialogue per role minimal. This enables LLMs to generate dialogues dynamically from configurations rather than memorizing specific roles, enhancing flexibility and adaptability. Additionally, we propose RoleRM in our benchmarks to address the challenge of evaluating RP performance.
+- **Interactive Engagement**: Measures the text's ability to draw the user in, encouraging ongoing interaction and contributing dynamically to the dialogue.
 
 
-# 3. Performance
-
-| Models               | Overall | Language Fluency | Language Relevance | Role Language | Role Knowledge | Emotional Expression | Interactive Engagement |
-|----------------------|---------|------------------|---------------------|----------------|-----------------|-----------------------|------------------------|
-| Llama-2-7B           | 1.57    | 2.19             | 1.83                | 1.63           | 1.37            | 1.21                  | 1.21                   |
-| Llama-3-8B           | 1.99    | 2.56             | 2.36                | 2.09           | 1.78            | 1.56                  | 1.60                   |
-| Llama-3.1-8B         | 1.94    | 2.52             | 2.30                | 2.01           | 1.75            | 1.47                  | 1.57                   |
-| Llama-2-7B-Crab      | 2.14    | 2.73             | 2.35                | 2.07           | 1.88            | 1.69                  | 2.12                   |
-| Llama-3-8B-Crab      | 2.22    | 2.81             | 2.51                | 2.16           | 1.95            | 1.77                  | 2.13                   |
-| **Llama-3.1-8B-Crab**| **2.23**| **2.87**         | **2.56**            | **2.17**       | **1.95**        | **1.76**              | **2.09**               |
-| GPT3.5               | 1.66    | 2.35             | 2.11                | 1.72           | 1.50            | 1.11                  | 1.17                   |
-| GPT4o                | 1.86    | 2.44             | 2.27                | 1.90           | 1.69            | 1.33                  | 1.51                   |
-| GPT4                 | 2.13    | 2.73             | 2.53                | 2.18           | 1.90            | 1.62                  | 1.86                   |
-| CharacterGLM-6B      | 1.83    | 2.37             | 1.96                | 1.80           | 1.60            | 1.39                  | 1.86                   |
-| Pygmalion-2-7B       | 2.11    | 2.82             | 2.49                | 2.01           | 1.86            | 1.58                  | 1.91                   |
-| Haruhi-Zero-7B       | 2.17    | 2.80             | 2.49                | 2.12           | 2.00            | 1.74                  | 1.86                   |
-
-
-Table 1: The results of evaluation on the test data of our Benchmark. The listed scores are from our RoleRM. Bold fonts indicate the best results and underlined fonts represent the second best. The subscripts represent the difference between each model and Crab (Llama-3.1-8B-Crab) counterpart.
 
+# 3. Performance
 
 
 <div align="center">
 
   <!-- Logo 图片 -->
-  <img src="https://cdn-uploads.huggingface.co/production/uploads/650add6348983c90ab688b6e/veerjA_MP5ZXmOxjAdZRO.png" width="500" style="border-radius: 20px;"/>
-
-</div>
-Figure 2: Human evaluation comparing Crab, GPT-3.5, and Pygmalion-2-7B. We selected a general LLM and one well-known RP-LLM to compare their generations against our Crab. For the same dialogue, annotators ranked responses from the three LLMs.
-
-
-| Models        | Overall | Language Fluency | Language Relevance | Role Language | Role Knowledge | Emotional Expression | Interactive Engagement |
-|---------------|---------|------------------|---------------------|----------------|-----------------|-----------------------|------------------------|
-| **Crab (sampled)** | **2.20** | **2.71**         | **2.45**            | **2.15**       | **1.95**        | **1.84**              | **2.12**               |
-| w/o base      | 2.17    | 2.72             | 2.41                | 2.07           | 1.89            | 1.79                  | 2.11                   |
-| w/o ref.      | 2.15    | 2.70             | 2.40                | 2.01           | 1.85            | 1.82                  | 2.11                   |
-| w/o scene     | 2.15    | 2.69             | 2.39                | 2.10           | 1.90            | 1.81                  | 1.98                   |
-
-Table 2: The ablation study for Crab. Due to missing attributes in our dataset, we sampled 1,000 fully attributed instances as the sub-test set to conduct the ablation experiments, referred to as Crab (sampled). The notation “w/o base" means without base role information for training RP-LLMs, including age, gender, personality, description, and expression; “w/o ref." means without catchphrases and knowledge; “w/o scene" means without interlocutor, relation, scenario, and tags.
-
-<br>
+  <img src="https://cdn-uploads.huggingface.co/production/uploads/650add6348983c90ab688b6e/22mf5uc1zpYLD0vGLtWM_.png" width="500" style="border-radius: 20px;"/>
 
-# 4. Three Datasets
-We publish three datasets, including Crab role-playing train set, Crab role-playing evaluation benchmark, and manually annotated role-playing evaluation dataset (can be used for training a Role-palying Evaluation Model).
+  Figure 1: The Spearman and Pearson correlations with human evaluations for the proposed RoleRM, ChatGPT, PairEval, G-Eval, and GPTScore. We average all aspects for calculations.
 
-## 4.1 Crab role-playing train set:
-{https://huggingface.co/datasets/HeAAAAA/Crab-role-playing-train-set}
+</div>
 
-## 4.2 Crab role-playing evaluation benchmark:
-{https://huggingface.co/datasets/HeAAAAA/Crab-role-playing-evaluation-benchmark}
 
-## 4.3 Crab manually annotated role-playing evaluation dataset:
-{https://huggingface.co/datasets/HeAAAAA/Crab-manually-annotated-role-playing-evaluation-dataset}
+<div align="center">
 
 
-<br>
+  <!-- Logo 图片 -->
+  <img src="https://cdn-uploads.huggingface.co/production/uploads/650add6348983c90ab688b6e/VKwol9kSa6qJPFGZ-B1jR.png" width="500" style="border-radius: 20px;"/>
 
-# 5. Fine-tuned Role-playing Model
-We release a fine-tuned model to achieve configurable Role-Playing tasks.
-{https://huggingface.co/HeAAAAA/Crab}
+  Figure 2: The comparison between RoleRM and ChatGPT. We calculate MAE to illustrate the gaps of Human Annotations with RoleRM and ChatGPT.
 
-<br>
+</div>
 
-# 6. Role-palying Evaluation Model
-We release a trained model to automate the evaluation of role-playing tasks.
-{https://huggingface.co/HeAAAAA/RoleRM}
 
-<br>
 
-# 7. Citation
+# 4. Citation
 
 ```bibtex
 @misc{kimiteam2025kimivltechnicalreport,