SPL: Orchestrating Workflows with Declarative Deterministic-Probabilistic Composition

2026-07-10T08:52:05Zbd983a1f34598a2f36ce7350d4320b9a7b40de75305d0e4a3ee1821eecf1f61c
ArduinoDWARFLLM orchestrationLLM-assisted-REPLCSPLbenchmarkingcode-modelsdecompilationfalse-positivesformal verificationgradient-estimationhardware-abstraction-layerindustrial control systemsinteger-overflowprobabilistic programmingprobabilistic-inferenceprovenancereverse-engineeringtype-probing

What happened

Collection of recent CS research with several security-relevant contributions. Key items: SPL (Structured Prompt Language) — a declarative orchestration language that composes deterministic solvers and probabilistic LLM calls, validated on a 78-recipe cookbook and a 1,200-run experiment showing solver-backed verification (82–93% verified correctness) versus unverified LLM output; REFORGE — a provenance-tracked pipeline and uncertainty-aware benchmark for LLM-assisted reverse engineering that quantifies how compiler optimization and alignment uncertainty reduce high-confidence ground truth (e.g

Why it matters

A reviewed impact interpretation has not been published for this record.

Evidence and limitations

Source ID
arxiv_cs_pl
Record identifier
bd983a1f34598a2f36ce7350d4320b9a7b40de75305d0e4a3ee1821eecf1f61c
Enrichment time
2026-07-10T08:52:05Z
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.