Newest papers

MRI Reconstruction Risk Estimation from Inference Dynamics

Tamir Shor, Alex Bronstein · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26558 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Rekonstruktor MRI iteratif menghasilkan perkiraan pertengahan sebelum mengembalikan gambar akhir. Kami mempelajari trajektori inferensi ini sebagai representasi model internal dari resiko rekonstruksi yang tercapai, diukur oleh log-NMSE. Status model-native diproyeksikan menjadi gambar magnitud, disimpulkan oleh deskriptor 16 dimensi tetap, dan dipaparkan ke risiko menggunakan bacaan pegunungan yang dilatih oleh sumber. Analisis kondisional menunjukkan bahwa informasi trajektorial dapat tetap terkait dengan kesalahan di luar prediktor yang hanya dimulai. Analisis prefix lebih lanjut menunjukkan bahwa informasi trajektorial yang berguna dapat muncul sebelum rekonstruksi selesai. Reproduksi kasus menyimpan asosiasi positif dan readout readout. Analisis kondisional menunjukkan bahwa informasi trajektorial dapat tetap terkait dengan kesalahan di luar prediktor yang hanya dimulai. Analisis

Machine translation (Meta M2M100); the English original follows.

Abstract

Iterative MRI reconstructors produce intermediate estimates before returning a final image. We study these inference trajectories as a model-internal representation of realized reconstruction risk, measured by log-NMSE. Model-native states are projected to magnitude images, summarized by a fixed 16-dimensional descriptor, and mapped to risk using a source-trained ridge readout. Across three reconstruction backbones, trajectory features are strongly predictive of reconstruction risk on held-out source data. Under zero-shot target shift, their behavior is backbone-dependent, exhibiting both preserved positive association and readout reversal. Conditional analyses show that trajectory information can remain associated with error beyond an initialization-only predictor. A prefix analysis further shows that useful trajectory information can emerge before reconstruction completes. Repeated-mask and low-shot calibration studies indicate that trajectories primarily encode persistent case difficulty, while their orientation and calibration can change under domain shift.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Trajectory, Inference, Computer science, Calibration, Artificial intelligence, Iterative reconstruction

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange