COMLLM uses multi-turn LLM reasoning via GRPO and LACS to achieve near-optimal latency, better fairness, and zero-shot generalization to larger unseen network topologies in mobile edge computing task offloading.
Serving long-context llms at the mobile edge: Test-time reinforcement learning-based model caching and inference offloading
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Multi-Turn Reasoning LLMs for Task Offloading in Mobile Edge Computing
COMLLM uses multi-turn LLM reasoning via GRPO and LACS to achieve near-optimal latency, better fairness, and zero-shot generalization to larger unseen network topologies in mobile edge computing task offloading.