MRM (Multi-Resolution Model): Partial Spoof Detector Checkpoint
Checkpoint for the IJCB 2026 submission "How Trustworthy Are Partial Spoof Detectors? A Cross-Domain Operational Audit."
What this is
A multi-resolution partial-spoof detector: the model of Zhang et al. (the
PartialSpoof multi-resolution countermeasure), via the open reimplementation
MultiResoModel-Simple (Luong et al.), trained by us on the PartialSpoof
training set. Front-end: wav2vec 2.0 Large; back-end: losses supervised jointly at
frame (20 ms), segment, and utterance scales.
This is not an authors'-released checkpoint of the original model and not
the public MultiResoModel-Simple checkpoint; it is our own training run.
File
| File | SHA256 |
|---|---|
55.pth |
5b753752f7c25370c6abf973f69f58e100dad4b5d3ea035872335358a876fdd1 |
Reported performance (PartialSpoof eval, ours)
- Utterance-level EER: 0.94%
- Segment-level (20 ms frame) EER: 13.91%
For reference, the public reimplementation checkpoint reports ~1.48% / ~13.67%; the original Zhang et al. model reports 0.49% utterance-level EER. Cross-domain behaviour (LlamaPartialSpoof, PartialEdit, HQ-MPSD) is the subject of the paper.
Training recipe
- Config: multi-resolution units {0.02, 0.04, 0.08, 0.16, 0.32, 0.64} s; segment
duration 9.6 s;
random_seek,use_mask; lr 1e-5; scheduler step 10, decay 0.5. - Trained to epoch 55. Not bit-reproducible (
random_seek, no fixed seed), which is why the trained weights are released directly.
Intended use
Research / reproducibility only. Audits an existing detector design under cross-domain partial-spoof attacks; not a deployable forensic tool.
License & attribution
Released under MIT, following the MultiResoModel-Simple reimplementation (MIT).
If you use this checkpoint, cite the original multi-resolution model (Zhang et al.,
IEEE/ACM TASLP 2023, The PartialSpoof Database and Countermeasures...) and the
reimplementation (Luong et al., ICASSP 2025, LlamaPartialSpoof), plus the IJCB
2026 paper.