dct

The article explains why tiny JPEGs can look different in Chrome compared to other browsers, due to a JPEG decoding optimization that skips high-frequency DCT coefficients during heavy downscaling.

Codec-Gauge learns small orthogonal channel transforms (gauges) for Transformer KV caches to improve compression fidelity at fixed bit rates, achieving significant KL divergence reduction across multiple models and backends.

Nigel Tao introduces Handsum, a new fixed-size LQIP image file format based on DCT, offering predictable file sizes and simplicity compared to JPEG.

CHIAR-Former uses spectral entropy-based routing to dynamically select between DCT, RBF, and self-attention operators, achieving improved efficiency on large text datasets while maintaining performance through hybrid attention mechanisms.