AnirbanSaha/mistral-nemo-12b-tlink-sentence-only-autoregressive
AnirbanSaha/mistral-nemo-12b-tlink-sentence-only-autoregressive is a 12.2 billion parameter model, fine-tuned from mistralai/Mistral-Nemo-Instruct-2407, specifically for temporal relation classification. It excels at identifying temporal links between sentences, classifying them as BEFORE, AFTER, OTHER, or NONE. This model achieves an accuracy of 0.7812 and a Macro-F1 of 0.7810 on its specialized task, making it suitable for applications requiring precise temporal ordering of events within text.
Loading preview...
Overview
AnirbanSaha/mistral-nemo-12b-tlink-sentence-only-autoregressive is a 12.2 billion parameter language model, fine-tuned from the mistralai/Mistral-Nemo-Instruct-2407 base model. Its primary function is to perform temporal relation classification between sentences, a specialized natural language processing task. This model is designed to identify and categorize the temporal relationship between two given sentences.
Key Capabilities
- Temporal Relation Classification: Accurately determines if one sentence occurs BEFORE, AFTER, or has an OTHER/NONE temporal relationship with another.
- Specialized Fine-tuning: Leverages the
AnirbanSaha/tlink-classification-sentence-onlydataset for its specific task, ensuring high relevance to temporal linking. - Robust Performance: Achieves an accuracy of 0.7812 and a Macro-F1 score of 0.7810 on its test set, with no invalid outputs recorded during evaluation.
Good For
- Event Ordering: Applications requiring the precise sequencing of events described in text.
- Timeline Generation: Building chronological timelines from unstructured textual data.
- Information Extraction: Extracting temporal dependencies for knowledge graphs or structured data.
- Document Analysis: Enhancing understanding of narrative flow and causality in documents.