Longformer lagged behind XLM‑RoBERTa on a 21‑label policy classification task across five languages, while GPT‑4 slightly outperformed the top open‑source model. Read more: https://getnews.me/evaluating-language-models-on-long-form-text-classification-tasks/ #longformtext #llm #policyclassification
September 17, 2025 at 4:38 AM
Longformer lagged behind XLM‑RoBERTa on a 21‑label policy classification task across five languages, while GPT‑4 slightly outperformed the top open‑source model. Read more: https://getnews.me/evaluating-language-models-on-long-form-text-classification-tasks/ #longformtext #llm #policyclassification