Metraipolis › archive › dc_da4172a7141a
PruneShift: A Framework for Evaluating Decision Reliability in Structured Pruning — arXiv:2608.29765
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.29765
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T08:36:54.813Z
- SHA-256 of the archived text
5fde7b9556638a92239e94e45a87f03c86c3869d9b6068ca1ccc7b90a3cbf653- Stable URL of this page
https://metraipolis.com/documento/dc_da4172a7141a
How to cite
PruneShift: A Framework for Evaluating Decision Reliability in Structured Pruning — arXiv:2608.29765 — arXiv (en), https://arxiv.org/abs/2608.29765 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_da4172a7141a (Metraipolis archive dc_da4172a7141a, sha256 5fde7b9556638a92239e94e45a87f03c86c3869d9b6068ca1ccc7b90a3cbf653)
BibTeX
@misc{dcda4172a7141a,
title = {PruneShift: A Framework for Evaluating Decision Reliability in Structured Pruning — arXiv:2608.29765},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.29765},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_da4172a7141a; sha256 5fde7b9556638a92239e94e45a87f03c86c3869d9b6068ca1ccc7b90a3cbf653},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.29765. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Structured pruning uses surrogate objectives because direct task evaluation over every feasible mask is too expensive. Most evaluations report average surrogate error or rank correlation on broadly sampled masks. These summaries do not directly test the mask chosen by the surrogate. We introduce PruneShift, an evaluation framework that separates broad predictive fidelity, fidelity near selector outputs, and the quality of the selected pruning decision. We first prove that Spearman and Kendall agreement can approach one while normalized selection regret remains maximal. We then derive sufficient conditions based on uniform error, selector suboptimality, decision margin, density ratio, and comparison mass. The analysis also yields a finite pool certificate with an explicit excess cost bound. Four studies test different links in this argument. External TextbookQA confirmation is heterogeneous: 7 of 20 simultaneous intervals favor the surrogate-selected mask, 6 favor its fixed comparator, and 7 cross zero. On a fixed Natural Questions pool, strict improvement holds in one of four settings. A controlled QQP experiment supports the proposed coverage mechanism in all 16 prespecified endpoints, although the sufficient bounds are conservative. Finally, a restricted OSSCAR reconstruction study on OPT-125M shows better local than broad fidelity in 68 of 75 primary endpoints. Independent fixed-mask confirmation is inconclusive in 24 of 25 endpoints and favors the comparator in one. These results show why predictive fit, decision reliability, and pruning method quality require separate evidence. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.LG). Categorie arXiv: Machine Learning (cs.LG) ; Neural and Evolutionary Computing (cs.NE). Pagina del lavoro: https://arxiv.org/abs/2608.29765 — 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 2026-10-17T08:13:26.766Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).