Multi-paradigm Logic Programming in the ${\cal E}$rgoAI System
2026-04-01T08:52:20Z•4d19cfaa8e45920a03d3b48b848918f93051ef2d634743bb0ea1535cd0977232
AgdaCon-DPOErgoAIFlora-2IoTPhyeldsPythonaggregate-computingarxivcode-generationcollision-fibersdependent-type-theoryerasureextractionfederated-learningformal-methodsknowledge-representationlogic-programmingpreference-learningprivacyrate-distortion theory','Lean4','confusability-graph','zero‑riskself-improvementsemantic-entropysemantic-identity-compressionvector-embeddings
What happened
This silver document aggregates several recent CS preprints: (1) ErgoAI — a multi-paradigm logic-programming language/system successor to Flora-2 focused on scalable KR&R, well-founded semantics, non‑monotonic inheritance and integration with external sources (e.g., vector embeddings); (2) ConSelf — a self-improving code-generation approach introducing code semantic entropy for curriculum construction and a consensus-weighted preference fine-tuning (Con-DPO) enabling improvement without superior teachers or test oracles; (3) a graded modal dependent type theory formalized in Agda that supports
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- arxiv_cs_pl
- Record identifier
- 4d19cfaa8e45920a03d3b48b848918f93051ef2d634743bb0ea1535cd0977232
- Enrichment time
- 2026-04-01T08:52:20Z
- 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.