Acceptance-Test-Driven Evaluation Protocols for Business-Centric LLM Systems

2026-06-03T08:51:46Z5d96e0f16532f41276e0adba41a3a0970cf734f3a1ad9497fc63752d7c40f2ec
Android developmentGNSS-IMULLM governanceLLM-assisted codingacceptance testingchange predictiondata minimizationembodied AIevaluation protocolsexecution-centered programmingformal verificationhuman-AI collaborationmobile sensingmulti-agent systemsorchestrationprivacyprogram synthesissoftware accountabilitysoftware engineering transformationsoftware qualitysoftware repairtesting

What happened

Collection of recent software-engineering research (arXiv, 2026-06-03) covering evaluation/governance for LLM-driven systems, accountability in software, privacy-by-design for Android (data minimization), orchestration and quality control for multi-agent software engineering (SPOQ), execution-centered program construction (AlgoTouch), models that relate code changes to runtime effects (Neural Change Prediction), the shifting role of engineers under Generative/Agentic AI, a multi-modal smartphone-based road-roughness assessment study, and a community agenda for reliable embodied AI (testing + (

Why it matters

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

Evidence and limitations

Source ID
arxiv_cs_se
Record identifier
5d96e0f16532f41276e0adba41a3a0970cf734f3a1ad9497fc63752d7c40f2ec
Enrichment time
2026-06-03T08:51:46Z
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.