Metraipolis › archive › dc_5b9bc2a7900c

Task-Oriented Rank Adaptation for Continual Learning in Text Classification — arXiv:2610.01702

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.01702
License (as archived)
CC0-1.0
Archived
2026-10-06T13:56:00.053Z
SHA-256 of the archived text
a7080ea2dfcc532442532ff2108cc7dd965a9e00713863f8113e96a59c77559f
Stable URL of this page
https://metraipolis.com/documento/dc_5b9bc2a7900c

How to cite

Task-Oriented Rank Adaptation for Continual Learning in Text Classification — arXiv:2610.01702 — arXiv (en), https://arxiv.org/abs/2610.01702 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5b9bc2a7900c (Metraipolis archive dc_5b9bc2a7900c, sha256 a7080ea2dfcc532442532ff2108cc7dd965a9e00713863f8113e96a59c77559f)
BibTeX
@misc{dc5b9bc2a7900c,
  title = {Task-Oriented Rank Adaptation for Continual Learning in Text Classification — arXiv:2610.01702},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.01702},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5b9bc2a7900c; sha256 a7080ea2dfcc532442532ff2108cc7dd965a9e00713863f8113e96a59c77559f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.01702. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Continual learning (CL) in text classification faces two critical challenges: catastrophic forgetting and negative transfer across sequential tasks. Parameter-Efficient Fine-Tuning (PEFT) methods such as LoRA enable efficient adaptation by learning low-rank updates of the model parameters. However, these compact representations are normally trained in isolation, limiting their reuse across related tasks. We introduce Task-Oriented Rank Adaptation (TORA), a geometric routing framework that leverages the low-rank structure of LoRA adapters to decide whether to transfer knowledge from the most compatible expert (Boosting) or isolate the new task (Shielding) based on structural similarity. Evaluated across 15 diverse text classification benchmarks, TORA consistently avoids harmful routing decisions: compatible tasks exceed their isolated performance while reducing training time, and structurally distant tasks are protected from interference with no loss in accuracy. With a single geometric threshold and no reliance on task identities or predefined sequences, TORA provides a simple and effective approach for dynamic adapter routing in sequential text classification systems.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.01702 — 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 2026-10-06T12:15:34.743Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).