{"ID":2894967,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.10468","arxiv_id":"2507.10468","title":"From BERT to Qwen: Hate Detection across architectures","abstract":"Online platforms struggle to curb hate speech without over-censoring legitimate discourse. Early bidirectional transformer encoders made big strides, but the arrival of ultra-large autoregressive LLMs promises deeper context-awareness. Whether this extra scale actually improves practical hate-speech detection on real-world text remains unverified. Our study puts this question to the test by benchmarking both model families, classic encoders and next-generation LLMs, on curated corpora of online interactions for hate-speech detection (Hate or No Hate).","short_abstract":"Online platforms struggle to curb hate speech without over-censoring legitimate discourse. Early bidirectional transformer encoders made big strides, but the arrival of ultra-large autoregressive LLMs promises deeper context-awareness. Whether this extra scale actually improves practical hate-speech detection on real-w...","url_abs":"https://arxiv.org/abs/2507.10468","url_pdf":"https://arxiv.org/pdf/2507.10468v1","authors":"[\"Ariadna Mon\",\"Saúl Fenollosa\",\"Jon Lecumberri\"]","published":"2025-07-14T16:46:30Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.LG\"]","methods":"[\"Transformer\",\"Large Language Model\"]","has_code":false}