R1-2407655
discussion
Discussion on AI/ML for CSI prediction
From Huawei
Summary
Huawei argues against introducing associated IDs for ensuring consistency between training and inference in CSI prediction for AI/ML models, proposing instead UE-side performance monitoring approaches. The document contains 2 proposals and 5 observations addressing feasibility, necessity, and alternative solutions.
Position
Huawei strongly advocates AGAINST introducing associated IDs or network-side indications for CSI prediction consistency, arguing they are unnecessary, technically infeasible, and create privacy risks. They push FOR UE-side performance monitoring as a sufficient implementation-based solution, differentiating CSI prediction from beam management where associated IDs were deemed necessary.
Key proposals
- Proposal 1 (Sec 2.2): No need to introduce indications from NW (e.g., explicit indication or associated ID) for ensuring consistency between training/inference for CSI prediction
- Proposal 2 (Sec 2.3): The consistency can be achieved by performance monitoring at UE side with an implementation manner