Metraipolis › archive › dc_2dc0354be249
Counterfactual Probing for Parallel Unmasking with Hidden Forest Structure — arXiv:2609.37841
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.37841
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T17:31:48.795Z
- SHA-256 of the archived text
7bd7a4074863ef7852cbe4ee34f38541c56a9b33cd31df0fb0867a0d5c0888ec- Stable URL of this page
https://metraipolis.com/documento/dc_2dc0354be249
How to cite
Counterfactual Probing for Parallel Unmasking with Hidden Forest Structure — arXiv:2609.37841 — arXiv (en), https://arxiv.org/abs/2609.37841 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2dc0354be249 (Metraipolis archive dc_2dc0354be249, sha256 7bd7a4074863ef7852cbe4ee34f38541c56a9b33cd31df0fb0867a0d5c0888ec)
BibTeX
@misc{dc2dc0354be249,
title = {Counterfactual Probing for Parallel Unmasking with Hidden Forest Structure — arXiv:2609.37841},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.37841},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2dc0354be249; sha256 7bd7a4074863ef7852cbe4ee34f38541c56a9b33cd31df0fb0867a0d5c0888ec},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.37841. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Masked generative models offer parallel token prediction, but accurate parallel sampling must account for dependencies among tokens. When dependencies are unknown, finding safe batches also costs model evaluations. We study whether total evaluations, including discovery, can be sublinear in sequence length $N$; sublinear sequential depth then follows. We consider discrete distributions with hidden forest structure, accessed through a fixed approximate conditional oracle. Under explicit regularity conditions and uniform Hellinger error bounds, for any fixed target accuracy $\varepsilon\in(0,1/8]$ and sufficiently large $N$, our sampler achieves seed-averaged total-variation error at most $\varepsilon$, with total masked-state submissions and sequential depth both bounded by $\widetilde{O}(N^C \varepsilon^{-a})$ for constants $0<C<1$ and $a>0$. These guarantees use polynomial vocabulary size and an edge-response lower bound set by $N$ and $\varepsilon$. The sampler shares evaluations of hypothetical reveals across dependence tests to identify safe parallel batches without requiring full recovery of the hidden forest. A tunable parameter trades probing cost against irreversible commit rounds. In the same class, any admissible irreversible product-commit sampler attaining the same seed-averaged accuracy requires $\Omega(N^c \varepsilon^b)$ counterfactual submissions or commit rounds in the worst case, for constants $c,b>0$.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2609.37841 — 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 2026-10-07T16:59:48.381Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).