Inference 0/0

Inference

Frontier inference

Recent inference research: FlashAttention, prefix sharing in a radix tree, multi-head latent attention, 4-bit floating-point formats and tree drafting. Each targets a cost from the earlier inference topics: memory traffic, cache size or decoding steps.

0 / 8 solved

Questions

0/8 solved
Question Tags Difficulty Success
1. What FlashAttention Does Instead of Storing S and P
flash-attentionbackward-pass +3
Hard —
2. The Critic Is Half the Memory
grporlhf +3
Hard —
3. Counting Nodes in a Draft Tree
speculative-decodingtree-attention +3
Hard —
4. KV Cache Paging
pagedattentionvllm +1
Hard —
5. What a Radix Tree Saves on a Shared Prompt
prefix-cachingradix-attention +3
Hard —
6. Latent KV, Decoupled RoPE
mlakv-cache-compression +1
Insane —
7. Counting Bits in NVFP4 and MXFP4
nvfp4mxfp4 +3
Hard —
8. Prefill/Decode Split
chunked-prefilldisaggregated-serving +1
Insane —