StickyInvoc introduces sticky tasks that load LLM model state once and invocation tasks that reuse it, yielding 3.6x speedup on a 150k-inference workflow.
Energy efficient protein language models: Leveraging small language models with lora for controllable protein generation,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
StickyInvoc: Rethinking Task Models for High-throughput Workflows in the LLM Era
StickyInvoc introduces sticky tasks that load LLM model state once and invocation tasks that reuse it, yielding 3.6x speedup on a 150k-inference workflow.