Metraipolis › archive › dc_c150eebf9ea4

Discriminative Flow Matching: Beyond Time-Conditioning in Generative Restoration via Flow-State Representations — arXiv:2609.04525

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.04525
License (as archived): CC0-1.0
Archived: 2026-10-17T00:43:26.568Z
SHA-256 of the archived text: 2a18e5945e1feae45a340ee9543c87a8e3c3bc00d17f29b1edd3a0b5aaefc75b
Stable URL of this page: https://metraipolis.com/documento/dc_c150eebf9ea4

How to cite

Discriminative Flow Matching: Beyond Time-Conditioning in Generative Restoration via Flow-State Representations — arXiv:2609.04525 — arXiv (en), https://arxiv.org/abs/2609.04525 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c150eebf9ea4 (Metraipolis archive dc_c150eebf9ea4, sha256 2a18e5945e1feae45a340ee9543c87a8e3c3bc00d17f29b1edd3a0b5aaefc75b)

BibTeX

@misc{dcc150eebf9ea4,
  title = {Discriminative Flow Matching: Beyond Time-Conditioning in Generative Restoration via Flow-State Representations — arXiv:2609.04525},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.04525},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c150eebf9ea4; sha256 2a18e5945e1feae45a340ee9543c87a8e3c3bc00d17f29b1edd3a0b5aaefc75b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.04525. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Existing Conditional Flow Matching (CFM) formulations describe transport progress using an explicit interpolation coordinate, commonly interpreted as time, assuming that a single global variable adequately represents a sample's position along the generative trajectory. In restoration tasks, however, transport progress is sample-dependent because the initial distribution may exhibit varying statistical dependencies with the target distribution. Thus, samples at the same interpolation coordinate can differ substantially in degradation level, distance to the target distribution, and restoration difficulty. We investigate whether signal representations learned by discriminatively trained models provide a meaningful description of generative transport state in CFM-based restoration. Through systematic latent-space analysis, we show that discriminative representations organize according to degradation severity and follow a consistent trajectory toward the clean-data manifold during generation. Motivated by these observations, we introduce the Discriminative Flow-State Hypothesis, which posits that discriminative representations encode a transport state governing generative restoration. Based on this hypothesis, we propose Discriminative Flow Matching, which conditions the Flow-Matching velocity field on Discriminative Flow-State Representations rather than explicit time coordinates. Experiments on speech enhancement and image denoising show that these representations characterize restoration progress, enable adaptive inference, and consistently outperform CFM and diffusion-related baselines. Our findings suggest that discriminative representations provide an effective state-aware alternative to explicit time conditioning and offer a novel perspective on the relationship between discriminative and CFM-based generative modeling.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-07 in eess.AS).
Categorie arXiv: Audio and Speech Processing (eess.AS) ; Sound (cs.SD).
Pagina del lavoro: https://arxiv.org/abs/2609.04525 — 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 2026-10-17T00:21:40.226Z dalla pagina https://arxiv.org/catchup/eess.AS/2026-09-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).