Donghoon Han

Conditional memory lets a language model add capacity through lookup, but the whole embedding table must stay resident; DeepSeek's table alone is 203 GB. Conditional Program stores each row as a short code plus a local scale and reconstructs it with a Gaussian trellis shared across the table, decoding only the rows an input requests. At two bits, storage falls roughly four- to eightfold, most reported task scores stay close to their originals, and output KL is 41–63% lower than 2-bit round-to-nearest.

Conditional memory lets a language model add capacity through lookup, but the whole embedding table must stay resident; DeepSeek's table alone is 203 GB. Conditional Program stores each row as a short code plus a local scale and reconstructs it with a Gaussian trellis shared across the table, decoding only the rows an input requests. At two bits, storage falls roughly four- to eightfold, most reported task scores stay close to their originals, and output KL is 41–63% lower than 2-bit round-to-nearest.