Metraipolis › archive › dc_ee7a8666199d

Beyond Fine-Tuning: Evaluating Prompt Engineering and Prompt Tuning for Roman Urdu Hate Speech Detection — arXiv:2611.04342

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.04342
License (as archived)
CC0-1.0
Archived
2026-11-04T21:37:54.415Z
SHA-256 of the archived text
5c79a64039e420ccd347ab0e3135442a8bf346ca77c97d7ce95015bf4fdfaf24
Stable URL of this page
https://metraipolis.com/documento/dc_ee7a8666199d

How to cite

Beyond Fine-Tuning: Evaluating Prompt Engineering and Prompt Tuning for Roman Urdu Hate Speech Detection — arXiv:2611.04342 — arXiv (en), https://arxiv.org/abs/2611.04342 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ee7a8666199d (Metraipolis archive dc_ee7a8666199d, sha256 5c79a64039e420ccd347ab0e3135442a8bf346ca77c97d7ce95015bf4fdfaf24)
BibTeX
@misc{dcee7a8666199d,
  title = {Beyond Fine-Tuning: Evaluating Prompt Engineering and Prompt Tuning for Roman Urdu Hate Speech Detection — arXiv:2611.04342},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.04342},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ee7a8666199d; sha256 5c79a64039e420ccd347ab0e3135442a8bf346ca77c97d7ce95015bf4fdfaf24},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04342. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Low Resource Languages (LRLs) pose significant challenges for hate speech detection because of a scarcity of readily available annotated data, language variation, non-standard spelling, and context-dependent ambiguity. Beyond the problem of informal conventions of writing in Roman Urdu, there is also the issue of code-mixing. After working on the previous project about zero-shot learning and parameter-efficient fine-tuning (PEFT), this research systematically compares prompt engineering and prompt tuning for Roman Urdu hate speech detection. To study the effect of in-context examples, we will compare the zero-shot and few-shot prompting approaches. To understand the advantage of structured prompting and learnable prompt representations, we will conduct studies of manual vs. mixed prompt-based adaptation strategies. Experimentations are carried out on the PURUT dataset (~72,771 social media comments) with GPT-3.5, LLaMA-3, Mistral-7B, and multilingual BERT in several settings. Experimental results show that few-shot prompting always outperforms zero-shot prompting, with the Mistral-7B model scoring 0.67 F1 compared to 0.56 of the zero-shot models. Based on this observation, prompt-based adaptation is proposed to introduce additional improvements, with the proposed mixed prompt configuration able to yield a F1-score up to 0.87 which further surpasses the performance of regular manual prompts. Despite the larger size of the model, its performance trade-off between inference-based prompting, prompt optimization, and parameter-efficient model adaptation is compared to the previously reported PEFT-based F1-score, which is ~0.93.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.04342 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-04T21:35:49.538Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).