CORTIVA Hits 73.5% Top-1 in EEG-to-Image Retrieval
CORTIVA, a candidate-score fusion framework for EEG- and MEG-to-image retrieval, reached 73.5% Top-1 and 95.3% Top-5 accuracy across ten participants on the 200-way THINGS-EEG2 benchmark, beating the strongest reported baseline by 10.3 and 5.4 percentage points. Instead of compressing heterogeneous visual supervision into a single embedding before ranking, the authors let separate decoding routes align to different visual targets and score the same candidate pool independently, merging only the temperature-scaled score vectors, which they say preserves complementary evidence; with a modality-specific encoder the same approach reached 42.4% Top-1 on THINGS-MEG. The work is a preprint and has not been peer reviewed.
Why it matters
The gain comes from where the fusion happens rather than from a bigger model, which suggests the ceiling on EEG-to-image retrieval is partly a ranking-architecture problem, and it is cheap enough for other groups to test before peer review settles the claim.