GeoRVQ: Decoder-aware geometry for residual-token prediction in physiological signals

arXiv:2609.27018v1 Announce Type: new Abstract: Residual vector quantization (RVQ) turns physiological waveforms into compact token sequences, but conventional masked modeling treats every incorrect token as equally costly. We propose GeoRVQ, a coarse-to-fine masked token model whose objective reflects the local response of a frozen waveform decoder. Decoder-induced costs define geometry-aware soft targets and expected distortion, while quantizer-causal prediction follows residual dependencies from coarse to fine levels. In a descriptive aggregate over MIMIC-IV Waveform, VitalDB, and CODE-15%, GeoRVQ increases exact token accuracy from .133pm.004 to .143pm.003, reduces decoded distance from .606pm.006 to .393pm.007, and increases R-peak F1 from .784pm.004 to .837pm.008 under matched model and training conditions. Across 45 held-out code substitutions, decoder-induced cost has a Spearman correlation of .85 with realized decoded cost, compared with .54 for Euclidean codeword distance. These results indicate that decoder-aware objectives can improve waveform and event preservation without requiring a large increase in exact token accuracy.

Source: arXiv cs.LG | 2026-09-24