Do Vision CIL Methods Transfer to Protein? A Diagnostic Benchmark of Frozen Protein Encoders

We conduct a controlled benchmark of nine representative CIL methods across five sequence-, structure-, and sequence--structure-based protein encoders and two enzyme function datasets at different scales, while keeping the encoders frozen and evaluation settings consistent across representations. It comes from a paper posted to arXiv on 20 October 2026. Protein function annotation operates over an expanding label space, making class-incremental learning (CIL) a natural framework for incorporating newly discovered functions without full retraining. Yet existing CIL methods have been developed largely in vision and have not been systematically evaluated over frozen protein representations. We find that no method transfers reliably across encoders, and that these failures persist across alternative phase constructions. Diagnostic analyses reveal that this instability arises from weak gradient-based stability signals, substantial heterogeneity in embedding geometry, and covariance assumptions whose appropriate capacity changes with both dataset scale and encoder. These findings suggest that the representation itself should be treated as a central component of continual-learning design rather than a passive feature extractor. More generally, reliable protein CIL will require encoder-aware methods that adapt their statistical assumptions and capacity to the geometry of the representation on which they operate. Code is available at https://github.com/zygao930/ProteinCIL.

Why it counts

We conduct a controlled benchmark of nine representative CIL methods across five sequence-, structure-, and sequence--structure-based protein encoders and two enzyme function datasets at different scales, while keeping the encoders frozen and evaluation settings consistent across representations. Code is available at https://github.com/zygao930/ProteinCIL.

Sources

Primary source: primary source

What is not known

This brief does not claim independent replication. Claims that appear only on X and not in the primary source stay unknown.