Closed-Loop Bayesian Bandit Encoder with GRAND Receiver for a Bursty Interference Channel

This batch contains diverse communications, coding, and hardware papers with several security-relevant implications. Key points: (1) A closed-loop Bayesian/GRAND receiver that adapts noise models and transmission mode (interleaved vs non-interleaved) — improves decoding but creates a learning transient and a dependence of performance on receiver-side model updates (potential attack surface for model manipulation or jamming during the transient). (2) BIHT 1-bit compressed sensing work shows that without per-iterate normalization iterates can oscillate under adversarial sign flips — a concrete,演