Separating Intelligence from Execution: A Workflow Engine for the Model Context Protocol
2026-05-05T08:52:25Z•e90105dd0874246a004bd91e95448b82b8c80dc61b81235be6528b957c57bd27
FAST-GEDGPU-accelerationGhostServeKV-cacheLLM-servingMCP (Model Context Protocol)SCIONSURGETezosblockchain-economicscache-orchestrationcheckpointingconfidentialitydata-leakageedge-computingembeddingserasure-codinggraph-edit-distanceliquid-stakingmediator-patternncsimresource-allocation','convex-optimization','CvxCluster'staking-protocolwireless-interferenceworkflow-engine
What happened
This is a multi-paper arXiv feed (May 2026) covering systems research: an MCP-native Workflow Engine and the MCP Mediator pattern for decoupling agent reasoning from execution; GhostServe, an erasure-coded in-host-memory checkpointing design to protect LLM KV caches; SURGE and FAST-GED for high-throughput GPU encoding and graph-edit-distance computation; Canonical LST (sTEZ), a protocol-native liquid-staking design for Tezos; SCION for size-aware cache policy orchestration; FPTC, a fast asymmetric signal codec; ncsim, a DAG scheduler + wireless-interference simulator for edge computing; a call
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- arxiv_cs_dc
- Record identifier
- e90105dd0874246a004bd91e95448b82b8c80dc61b81235be6528b957c57bd27
- Enrichment time
- 2026-05-05T08:52:25Z
- 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.