Symposium proceedings
Proceedings of the Symposium on Efficient and On-Device Machine Learning
SEODML 2026
- 8 September 2026 – 10 September 2026
- Singapore
- Kestrel AI Institute
Sample volume. The event, organisers and papers are fictional demonstration content.
About this volume
SEODML focuses on making machine-learning systems small, fast and energy-efficient enough to run where the data is produced. Topics include quantisation, distillation, adapters, memory-bounded inference and measurement on real devices.
SEODML-2026-proceedings.pdf
Papers (5)
Quantisation and distillation
- pp. 1–10360
Full paperBest paper
Mixed-Precision Distillation for 4-bit Phone-Class ModelsYuki Saito, Daniel Okafor
10.xxxx/ctm.pr.2026.0006
downloads - pp. 27–32140
Short paper
Low-Rank Adapters for On-Device PersonalisationPriya Venkataraman, Anders Holm
10.xxxx/ctm.pr.2026.0009
downloads
On-device inference
- pp. 11–20280
Full paper
KV-Cache Compression Under Strict Memory BudgetsRahul Desai, Anders Holm
10.xxxx/ctm.pr.2026.0007
downloads - pp. 21–26175
Short paper
Profiling Energy Use of Small Language Models on Mobile SoCsYuki Saito, Mei Tanaka
10.xxxx/ctm.pr.2026.0008
downloads - pp. 33–3496
Poster
Benchmarking Sub-Billion-Parameter Models on Commodity HardwareMei Tanaka, Daniel Okafor
10.xxxx/ctm.pr.2026.0010
downloads