Metraipolis › archive › dc_c7fd36c9775e

Train Smarter, Not Harder: Switching Signal-Guided Training in Active Learning — arXiv:2609.06806

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.06806
License (as archived)
CC0-1.0
Archived
2026-10-16T12:56:43.800Z
SHA-256 of the archived text
5358a94c9301b6b0be2260280c1ddb12b0a9808d94fcbd2e8593c563d82396f2
Stable URL of this page
https://metraipolis.com/documento/dc_c7fd36c9775e

How to cite

Train Smarter, Not Harder: Switching Signal-Guided Training in Active Learning — arXiv:2609.06806 — arXiv (en), https://arxiv.org/abs/2609.06806 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c7fd36c9775e (Metraipolis archive dc_c7fd36c9775e, sha256 5358a94c9301b6b0be2260280c1ddb12b0a9808d94fcbd2e8593c563d82396f2)
BibTeX
@misc{dcc7fd36c9775e,
  title = {Train Smarter, Not Harder: Switching Signal-Guided Training in Active Learning — arXiv:2609.06806},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.06806},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c7fd36c9775e; sha256 5358a94c9301b6b0be2260280c1ddb12b0a9808d94fcbd2e8593c563d82396f2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.06806. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Training strategy, namely whether to retrain from scratch or fine-tune from the previous checkpoint, is an overlooked decision variable in active learning. We show that this choice has exploitable structure: retraining is most useful in early rounds, when each batch can substantially reshape the labeled distribution, while fine-tuning becomes safer once the model trajectory stabilizes. We propose HybridAL, an adaptive training schedule that monitors an online stabilization signal and switches from retraining to fine-tuning after sustained stabilization. Two complementary signals, spectral exponent change $\Delta\alpha$ (weight-based) and accuracy change $\Delta$Acc (validation-based), span different points on the time-calibration trade-off. Across three encoder backbones and six text-classification tasks (five seeds each), HybridAL keeps endpoint macro-F1 non-inferior to retraining and fine-tuning at a 0.010 margin, saves up to 49% of retraining time, and recovers a substantial fraction of retraining's calibration advantage as measured by negative log-likelihood (NLL). Compared with schedules that switch at a pre-committed round, HybridAL obtains lower NLL at moderate additional cost, showing that trajectory-dependent switching provides a stronger time-calibration trade-off than fixed early switching.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CL).
Categorie arXiv: Machine Learning (cs.LG) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.06806 — 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 2026-10-16T12:30:25.288Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).