Meta-Chunking: Learning Text Segmentation and Semantic Completion via Logical Perception
The summary and rewrite models were fully fine-tuned on the Qwen2.5-3B-Instruct utilizing 20K data entries from the CRUD benchmark, which was prepared with ERNIE-3.5-128K and QwQ-32B.
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support