Harald Koestler
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
Distributed×1AI×1ML×1
Frequent co-authors
Research Timeline
2026
Leyline: KV Cache Directives for Agentic Inference
Leyline introduces a novel serving-side primitive that allows agentic LLMs to perform targeted, efficient edits to the KV cache, avoiding costly full re-prefilling after content modification.
Highlighted terms show continued research focus across papers