EvoOntology: open-source self-evolving ontology layer bridges data agents' gap to heterogeneous data
TheTuringPost · x · 2026-09-26
Researchers from Renmin University open-sourced EvoOntology (paper, code on GitHub) to address the "agent-data gap": heterogeneous data (tables, files, databases) lives outside the agent, and neither raw exploration nor manually built semantic layers scale well.
- The ontology is packaged as an MCP server with a schema layer, content layer, and tool layer, letting agents query it at runtime.
- A builder agent constructs the ontology autonomously; a self-evolution loop refines it via attribution-guided typed edits accepted only after backbone-conditional paired evaluation.
- Across three widely used data-agent benchmarks and four LLM backbones, EvoOntology consistently beats strong baselines and existing semantic-layer approaches.
Related event: RUC Open-Sources EvoOntology for Data Agents(3 posts)→
More from coding & agent
- Calesto launches: a to-do app that delegates tasks to Claude Code and Codex — thesaraharminta · 2026-09-26
- Dev builds full pixel RPG with Opus 5.5 and Unity CLI — AI even edited the demo video — noahsolomon · 2026-09-26
- Microsoft open-sources Fabric-RLM: LLMs write code to recursively chew through big data — adnan_hashmi · 2026-09-26
- The Future of MCP: 6 Shifts for Production-Grade AI Agents — MaryamMiradi · 2026-09-26
- kaigen3d principles: code-first runtime and user-controlled frame loop make it AI-friendly — gdechichi · 2026-09-26
- kaigen3d engine demo: cross-platform AI-first 3D engine hits 330+ FPS on 6-year-old laptop GPU — gdechichi · 2026-09-26