The Journal of Advanced Computing and Intelligent Systems (JACIS) invites submissions to a special issue on Efficient Large Language Models at the Edge. As foundation models move from the data center to phones, vehicles, robots and industrial controllers, the trade-off between capability, latency, memory footprint and energy budget has become one of the defining engineering problems of the field. This issue aims to gather the most current work on making large models practical where compute is scarce.
Scope and Topics of Interest
We welcome original research, systems papers and rigorous benchmark studies on topics including, but not limited to:
- Model compression: quantization, pruning, knowledge distillation and low-rank adaptation
- On-device and on-nominal-hardware inference: scheduling, memory management and runtime design
- Federated and privacy-preserving fine-tuning on edge fleets
- Energy-efficient training and serving; hardware-aware neural architecture design
- Retrieval-augmented and cache-augmented generation under constrained bandwidth
- Evaluation methodology: edge-specific benchmarks, latency/quality trade-off metrics
- Applications: embedded AI, mobile assistants, in-vehicle intelligence, industrial monitoring
Important Dates
- Submission deadline: December 15, 2026
- First review decision: January 30, 2027
- Revised manuscripts due: February 28, 2027
- Target publication: April 2027 (JACIS special issue)
What We Offer
Consistent with ChainKnow Press's editorial model, every submission to this issue receives a fast editorial pre-check (within 48 hours), double-blind review by active researchers, and a decision target of four weeks. The outstanding contribution will be featured as the issue's cover article, with a cover story published on our website.
Submission Guidelines
Manuscripts must be prepared following the JACIS author guidelines and submitted through the journal platform. An ORCID iD is mandatory for the corresponding author. Submissions that are under review elsewhere will not be considered. There is no page limit, but authors are encouraged to keep the manuscript focused.
Contact
Questions about scope or fit may be directed to the JACIS editorial office at editor@chainknowpub.com. Please use the subject prefix [JACIS-EdgeSI].