viamr-project/qwen3-1.7b-amr-20260704-0113
The viamr-project/qwen3-1.7b-amr-20260704-0113 is a 2 billion parameter model with a 32768 token context length, developed by viamr-project. This model is specifically evaluated for its performance on AMR (Abstract Meaning Representation) tasks, achieving an average F1 score of 76.97. It is designed for applications requiring robust semantic parsing and understanding, making it suitable for tasks like information extraction and natural language understanding.
Loading preview...
Model Overview
The viamr-project/qwen3-1.7b-amr-20260704-0113 is a 2 billion parameter language model developed by viamr-project, featuring a substantial context length of 32768 tokens. This model has been specifically evaluated for its capabilities in Abstract Meaning Representation (AMR) tasks.
Key Capabilities
- Abstract Meaning Representation (AMR): The model demonstrates proficiency in AMR parsing, achieving an average F1 score of 76.97 across 1898 evaluated samples. It also shows an average precision of 76.65 and an average recall of 77.91.
- Semantic Parsing: Its strong performance in AMR indicates a robust ability to extract and represent the semantic structure of sentences.
- Large Context Window: With a 32768 token context length, it can process and understand longer texts, which is beneficial for complex semantic analysis.
Good For
- Semantic Understanding Applications: Ideal for use cases requiring deep linguistic analysis and the conversion of natural language into structured meaning representations.
- Information Extraction: Its AMR capabilities can be leveraged for precise information extraction from unstructured text.
- Research in NLP: Suitable for researchers exploring advanced semantic parsing and graph-based meaning representations.