arXiv:2606. 01502v1 Announce Type: cross Abstract: Frontier LLMs increasingly decide what a query attends to with a sparse-attention indexer that picks a few KV-cache blocks per query: attention's unit is now a small, reusable chunk.
arXiv:2609.03949v2 Announce Type: replace-cross Abstract: A long-lived KV cache must be compressed before the queries that will read it exist. Selection by observed attention collapses there: on a No...
arXiv:2606. 21633v2 Announce Type: replace Abstract: The KV cache dominates GPU memory in long-context LLM serving, crowding out batch capacity and leaving GPU compute idle.
arXiv:2609.14507v1 Announce Type: cross Abstract: Single-GPU long-context inference with Mixture-of-Experts (MoE) models requires spilling the key-value cache (KVCache) to CPU memory. The spilled KV...
arXiv:2604. 26968v2 Announce Type: replace-cross Abstract: Key-value (KV) cache memory management is the primary bottleneck limiting throughput and cost-efficiency in large-scale GPU inference serving.
VestigeKV is a new KV‑cache technique that uses a 64‑dimensional vestigial branch—originally a RoPE component repurposed during NoPE training—as a query‑independent eviction signal. By reading only 11 % of each cache row, the method partitions the cache into an attended tier (top‑m rows) and an archive tier (all other rows), which is GPU‑resident and never deleted. The approach achieves near‑perfect retrieval (1.00 at 8×, 0.92 at 32×) without any training, quantization, or changes to weights or kernels.
arXiv:2609.37626v1 Announce Type: cross Abstract: No single way of parallelizing attention serves large language models well under all loads. Low concurrency favors tensor parallelism, many independe...
The paper investigates how block‑diffusion language models can use a constant‑size cache to enable efficient parallel decoding. By employing sequence mixers that summarize completed blocks into a reusable state and a block‑causal training objective, the authors pretrain three 3B block‑diffusion denoisers (attention, Mamba, and hybrid) on 300 B tokens. The resulting state‑space cache remains O(1) in memory and latency regardless of context length, yielding significant speed‑up and memory savings compared to traditional attention‑based caches, especially at very long sequences.
arXiv:2606. 13361v1 Announce Type: new Abstract: Right now, across the world, AI agents are repeating the same absurd act: to read one document, they each recompute it from scratch.
arXiv:2606. 30709v1 Announce Type: cross Abstract: Hierarchical Global Attention (HGA) is a drop-in replacement for dense causal attention in pretrained long-context transformers.
arXiv:2607. 05876v1 Announce Type: cross Abstract: LLM serving optimization typically benchmarks many configurations and reaches for heavy profilers when latency targets are missed.
The paper investigates whether training Mixture-of-Experts (MoE) routers can improve memory‑bandwidth locality on consumer GPUs. Using a new zero‑surgery telemetry tool, the authors measure that a large Qwen3‑235B model is bottlenecked by disk‑based expert access, and that an LRU cache can serve a majority of requests. They pre‑register experiments training 137 M‑parameter MoE models with locality‑aware losses, finding that while cache misses can drop up to 60 % (99 % static‑pin hit rate), every configuration fails to meet a strict 1 % perplexity threshold, indicating a tight coupling between cache efficiency and model quality.
Serving a 235B-parameter Mixture-of-Experts (MoE) model on a single 8 GB GPU is bottlenecked not by compute but by memory bandwidth: decode must stream each token's active experts from whichever tier holds them, and on consumer hardware most experts sit on an SSD far slower than RAM. We quantify this bandwidth wall on Qwen3-235B (Q4_K_M, 134 GB): measured decode is 0.