Claim-to-evidence verification for AI answers and RAG: provenance, unsupported-claim detection, and reversible review. / AI 回答与 RAG 证据核验。
-
Updated
Aug 25, 2026
Claim-to-evidence verification for AI answers and RAG: provenance, unsupported-claim detection, and reversible review. / AI 回答与 RAG 证据核验。
AI-agent black-box recorder for tamper-evident DIKWP traces, incident replay, provenance, and audit. / AI Agent 黑匣子与可审计语义追踪。
Offline functional white-box evaluation for LLMs across DIKWP transformations, inspectable artifacts, risks, and boundaries. / 大模型功能白盒评测。
Offline research studio for testable artificial-consciousness indicators, claim levels, evidence ledgers, scenarios, and governance. / 人工意识评测与治理。
Compile mathematical problems into deterministic, hash-linked, machine-readable audit certificates with explicit proof boundaries. / 语义数学问题—证书编译器。
Offline-first K–12 continuity and autonomous-learning OS for uncertain AI/AGI futures: rights, agency, resilience, and reversible pathways. / 学生成长韧性系统。
Official GitHub research hub for Yucong Duan (段玉聪): DIKWP, artificial consciousness, semantic mathematics, auditable AI, patents, conferences, and open-source systems.
Add a description, image, and links to the dikwp topic page so that developers can more easily learn about it.
To associate your repository with the dikwp topic, visit your repo's landing page and select "manage topics."