All Research tracks

AI Science & Algorithms

The algorithms behind a model's answer — attention, the KV cache, speculative decoding, quantization, and vector recall — derived properly, to where each one breaks.

1 series · 4 articles

LLM Inference Math

1Attention and the KV Cache: The Arithmetic of a Forward PassWatch

2Speculative Decoding: The Accept-Reject Math of Faster TokensWatch

3Quantization Error: How Low Can the Bits Go Before Quality BreaksWatch

4Vector Indexes: The Recall-vs-Latency Trade-off of HNSW and IVFWatch