WorldKV: managing a world model's memory like a cacheMemory in autoregressive video world models is a bind: full KV attention preserves revisit consistency, but memory footprint and attention cost grow linearly with rollout length, passing 200 GB per minute and exceeding a B200. Sliding window is real-time but drifts. WorldKV trains nothing and works purely at inference: it retrieves evicted KV chunks by camera/action similarity and compresses each chunk to half size by key-key cosine similarity, fitting roughly twice the history at roughly twice the throughput of full KV, with no memory training at all.Jung Yi, Minjae Kim, Paul Hyunbin Cho · 578 views · May 21, 2026WorldModelKVCacheMay 21, 2026
0:14WorldKV: retrieval and compression for world-model memory@jungyi3578 · 429 views · 2026-05-23WorldModelKVCache
0:15DeepGEMM-Ascend Brings DeepSeek Kernels to Huawei NPUs@EngMoElgaraihy · 521 views · 2026-10-05DeepGEMMDeepGEMM-Ascend