Agricultural named entity recognition technology based on thought chain distillation and counterfactual reasoning
ObjectiveTo address the issues of hallucinations, contextual logical inconsistencies, and inability to run on low-resource devices when large language models perform named entity recognition (NER) in agriculture.MethodUsing DeepSeek with 671 billion parameters (DeepSeek-671B) as the teacher model, d...
Αποθηκεύτηκε σε:
| Κύριοι συγγραφείς: | , , , , , |
|---|---|
| Μορφή: | Artigo |
| Γλώσσα: | Chinês |
| Έκδοση: |
South China Agricultural University
2026-01-01
|
| Σειρά: | Huanan Nongye Daxue xuebao |
| Θέματα: | |
| Διαθέσιμο Online: | https://journal.scau.edu.cn/article/doi/10.7671/j.issn.1001-411X.202507003 |
| Ετικέτες: |
Δεν υπάρχουν, Καταχωρήστε ετικέτα πρώτοι!
|