← back to paper

Review history

arxiv: 2606.06256 · 2 revisions

RedKnot: Efficient Long-Context LLM Serving with Head-Aware KV Reuse and SegPagedAttention

1. 2026-06-29 UNVERDICTED LOW v0.9.1-grok novelty 6.0
   24261 ms 5853 in 1074 out 2026-06-29T05:05:13.850884+00:00
2. 2026-06-28 UNVERDICTED LOW v0.9.1-grok novelty 6.0
   33549 ms 5853 in 1213 out 2026-06-28T00:58:12.523350+00:00