arXiv Machine Learning
Fathom: Per-Query Read Depth for Sparse Decoding over Offloaded KV Caches
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv Machine Learning
Sep 3
Unfolding the Leech Lattice: Fused Multi-Shell Decoding and VRAM Layouts for 2-Bit LLM Weights
The paper introduces a multi‑shell decoder for Leech‑lattice vector quantization, achieving the best reported 2‑bit quality under its evaluation protocol. It presents a GPU‑friendly layout that fuses dequantization with matrix‑vector multiplication, demonstrating significant speed and memory advantages over traditional one‑hot masks and other 4‑bit methods. Experiments show the new kernel outperforms baseline approaches across multiple model sizes, with measurable gains in throughput and reduced byte traffic.
arXiv Machine Learning
Aug 11
SPECTRA: Pushing the KV Cache Beyond the 2-Bit Cliff via Spectral Transform Coding
arXiv:2608. 07915v1 Announce Type: new Abstract: Large language models (LLMs) increasingly read long inputs in the agentic era, from whole documents and codebases to conversations across many turns.
arXiv Machine Learning
Jul 9
Fractal KV-Cache Archives: Lossless Symbolic Storage with In-Place Retrieval for Long-Context LLM Inference
arXiv:2607. 07144v1 Announce Type: new Abstract: The key-value (KV) cache dominates the memory cost of long-context autoregressive inference, and a growing body of work compresses it through quantization, eviction, or offloading.
arXiv AI
Jun 19
UltraQuant: 4-bit KV Caching for Context-Heavy Agents
arXiv:2606. 20474v1 Announce Type: cross Abstract: Context-heavy agents place unusual pressure on the key-value (KV) cache: long prefixes are reused across many short turns, while concurrency determines whether the serving system can keep GPUs utilized.
arXiv Machine Learning
Aug 4
HERALD: High-Throughput Block Diffusion LLM Serving via CPU-GPU Cooperative KV Cache Retrieval
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 Computation and Language
Sep 10
VestigeKV: The NoPE-MLA KV Cache Carries Its Own Sparse-Attention Signal in a Vestigial Branch
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...