Metraipolis › archive › dc_6e49ca99cef7
Asymmetric Within-Document Predictive Learning for Scientific Document Representation — arXiv:2608.28625
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.28625
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T07:47:08.048Z
- SHA-256 of the archived text
fd396af8ac39cc63492b8b7a2906d7a3f9f5ed394c9bbd2d9c254940e0449369- Stable URL of this page
https://metraipolis.com/documento/dc_6e49ca99cef7
How to cite
Asymmetric Within-Document Predictive Learning for Scientific Document Representation — arXiv:2608.28625 — arXiv (en), https://arxiv.org/abs/2608.28625 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6e49ca99cef7 (Metraipolis archive dc_6e49ca99cef7, sha256 fd396af8ac39cc63492b8b7a2906d7a3f9f5ed394c9bbd2d9c254940e0449369)
BibTeX
@misc{dc6e49ca99cef7,
title = {Asymmetric Within-Document Predictive Learning for Scientific Document Representation — arXiv:2608.28625},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.28625},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6e49ca99cef7; sha256 fd396af8ac39cc63492b8b7a2906d7a3f9f5ed394c9bbd2d9c254940e0449369},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.28625. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We study predictive pretraining for scientific document representation using the discourse structure of papers. We propose SciJEPA, a citation-free framework that learns through asymmetric within-document prediction: title and abstract representations are used to predict method representations, and method representations are used to predict conclusion representations. Experiments on RELISH, high-influence citation, SciDocs, and cite prediction show that plain predictive training is viable but weaker than a controlled contrastive baseline using the same section pairs. Adding Sliced Isotropic Gaussian Regularization (SIGReg) substantially improves performance and narrows this gap. The effect of regularization is task-dependent: moderate SIGReg helps fine-grained ranking, while stronger regularization can weaken local alignment. We further show that different encoding branches support different retrieval regimes. These results position within-document predictive learning as a promising citation-free complement for scientific document representation, provided that embedding geometry is carefully controlled. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.28625 — 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 2026-10-17T07:28:26.664Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).