RSH-SpMM: A Row-Structured Hybrid Kernel for Sparse Matrix-Matrix Multiplication on GPUs
2026-03-11T08:52:37Z•964c0ff3a83bb55b20459c4dcc833f0142fb32359c526e775d3b41067847541d
AuralinkSDCFedAvgFedLECCISO-15118LLM-fine-tuningOCPPPREEMPT_RTPagedAttentionQLoRASCAFFOLDZipage','kv-cache-eviction'autonomous-systemscompressed-kv-cachedeZentdecentralized-privacyedge-aiev-chargingfederated-learningmodel-inversionnon-iid-datapoisoning-resilienceprivacysecure-sumstochastic-countingz-anonymity
What happened
This collection of arXiv papers (Mar 11 2026) presents multiple systems and algorithms with direct security, privacy, and safety implications. Notable items: Auralink SDC proposes edge-deployed autonomous LLM agents for EV charging management (uses models fine-tuned on OCPP/ISO15118 and incident logs) — a safety-critical attack surface where exploitation or erroneous autonomous remediation could cause large-scale service disruption or unsafe actions. Federated learning papers (benchmarking and FedLECC) highlight client-selection, non-IID challenges, and gaps in privacy/robustness, raising risk
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- arxiv_cs_dc
- Record identifier
- 964c0ff3a83bb55b20459c4dcc833f0142fb32359c526e775d3b41067847541d
- Enrichment time
- 2026-03-11T08:52:37Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.