Metraipolis › archive › dc_ef3dcbc8863c

When Links Mislead: Label-Free Structural Polarity for Zero-Shot Classification of Web Pages — arXiv:2610.41464

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.41464
License (as archived)
CC0-1.0
Archived
2026-11-03T08:11:55.060Z
SHA-256 of the archived text
fe55350957561e0cddf01e2d292f9278344c49a8ada70e8405ca5099d17ca123
Stable URL of this page
https://metraipolis.com/documento/dc_ef3dcbc8863c

How to cite

When Links Mislead: Label-Free Structural Polarity for Zero-Shot Classification of Web Pages — arXiv:2610.41464 — arXiv (en), https://arxiv.org/abs/2610.41464 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ef3dcbc8863c (Metraipolis archive dc_ef3dcbc8863c, sha256 fe55350957561e0cddf01e2d292f9278344c49a8ada70e8405ca5099d17ca123)
BibTeX
@misc{dcef3dcbc8863c,
  title = {When Links Mislead: Label-Free Structural Polarity for Zero-Shot Classification of Web Pages — arXiv:2610.41464},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41464},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ef3dcbc8863c; sha256 fe55350957561e0cddf01e2d292f9278344c49a8ada70e8405ca5099d17ca123},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41464. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Web platforms such as digital libraries, online stores and university sites must classify new pages before anyone labels them. Large language models (LLMs) can label each page from its text, but these labels are noisy, and links could correct them only where linked pages share classes. Existing zero-shot methods ignore the links, or pull a page toward its neighbours as label propagation and graph neural networks (GNNs) do, which hurts where linked pages differ in class. Others weight propagation by an LLM-estimated homophily that costs extra calls and drifts across LLMs. We find that the sign with which links should enter a prediction, which we call structural polarity, is readable without true labels. It is the excess agreement of LLM labels and text-prototype predictions along links. We thus propose ZeroPolar, which gates one signed propagation step by this statistic. Training-free ZeroPolar mixes prototype scores with any number of LLM labels, and ZeroPolar-T denoises the labels with a text classifier once half of the pages carry one. With one GPT-4o label per page, ZeroPolar-T reaches the highest mean accuracy on eleven graphs (73.99%), 6.01 points above LLM-BP and significantly above all eighteen baselines we compare. With LLM labels on only 5% of the pages, training-free ZeroPolar leads a GNN trained on these labels by 10.80 points.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.41464 — 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 2026-11-03T08:02:38.926Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).