LLM Digest

Story

arxiv_cs_lg · Oct 28, 2026 · paper

Source brief

Causal Attention on CPU: A Cache-Aware and Load-Balanced Scheduling Kernel for Prefill

arxiv.orgOct 28, 2026
original source linked

In brief

Agentic AI drives significant productivity gains across diverse domains, but concurrent long-context prefills from multiple agents can saturate GPU compute resources. Although CPUs offer a viable alternative, existing...

Feed lens

agentic

Read the original at arxiv.org →Open in live feed Read that day’s brief

Causal Attention on CPU: A Cache-Aware and Load-Balanced Scheduling Kernel for Prefill

Earlier in this thread 4 items

Uber Moves beyond CPU Cores with Performance-Aware Scheduling for Stateful Workloads

Causal Chain of Thought: Mitigating Attention Leakage via Context Isolation

PhaseGate: Phase-Aware CPU Retrieval Scheduling for On-Device LLMs on Unified Memory

Heads of Security: Learned Attention Gates Improve Prompt Injection Robustness