{"slug": "teachmategpt-a-multi-agent-knowledge-grounded-framework-for-pedagogical-from", "title": "TeachMateGPT: A Multi-Agent Knowledge-Grounded Framework for Pedagogical Assessment Generation from Science Curriculum Materials", "summary": "TeachMateGPT, a multi-agent framework introduced in an arXiv paper (arXiv:2608.13708v1), improves curriculum-grounded science assessment generation by raising faithfulness from 0.68 to 0.96 and answer relevancy from 0.60 to 0.89 over a vanilla RAG baseline. The system introduces COPE, a hierarchical knowledge base; a staged fail-closed agent pipeline; SAVER, a source-attributed verification protocol; and NCTB-SciGen8, a dataset of 198 items spanning all 14 chapters of the NCTB Class 8 science textbook, rated by three practicing teachers.", "body_md": "arXiv:2608.13708v1 Announce Type: new\nAbstract: Automatically generating textbook-grounded assessment items can reduce science teachers' workload, but existing retrieval-augmented generation (RAG) systems rely on flat retrieval, support only single-question generation, lack safeguards against weak evidence, and are ill-suited to low-resource, board-exam-structured curricula. We address these limitations with TeachMateGPT, a multi-agent system contributing four advances to curriculum-grounded science-assessment authoring. (i) COPE, a hierarchical knowledge base replacing token-window chunking with a multi-resolution index that segments documents along syllabus structure and links them at three granularities via a traversable graph-based lineage, matching evidence to each topic's instructional level. (ii) A staged, fail-closed agent pipeline replacing one-shot retrieve-then-generate: routing gates search, retrieval fuses dense and lexical evidence under a coverage gate that withholds generation on insufficient evidence, and specialist agents draft objective and constructed-response items. (iii) SAVER, a source-attributed verification protocol scoring faithfulness, relevance, and hallucination risk against retrieved evidence, applying stricter grounding checks across each creative question's four sub-parts, paired with teacher-in-the-loop evaluation rather than automatic filtering. (iv) NCTB-SciGen8, a curriculum-grounded dataset of 198 items (143 multiple-choice, 55 creative questions) spanning all 14 chapters of the NCTB Class 8 science textbook, produced by the pipeline and rated by three practicing teachers. TeachMateGPT raises faithfulness (0.68 $\\rightarrow$ 0.96) and answer relevancy (0.60 $\\rightarrow$ 0.89) over a vanilla RAG baseline.", "url": "https://wpnews.pro/news/teachmategpt-a-multi-agent-knowledge-grounded-framework-for-pedagogical-from", "canonical_source": "https://arxiv.org/abs/2608.13708", "published_at": "2026-08-17 04:00:00+00:00", "updated_at": "2026-08-17 04:14:10.490145+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "natural-language-processing", "ai-agents", "ai-research"], "entities": ["TeachMateGPT", "COPE", "SAVER", "NCTB-SciGen8", "NCTB Class 8 science textbook", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/teachmategpt-a-multi-agent-knowledge-grounded-framework-for-pedagogical-from", "markdown": "https://wpnews.pro/news/teachmategpt-a-multi-agent-knowledge-grounded-framework-for-pedagogical-from.md", "text": "https://wpnews.pro/news/teachmategpt-a-multi-agent-knowledge-grounded-framework-for-pedagogical-from.txt", "jsonld": "https://wpnews.pro/news/teachmategpt-a-multi-agent-knowledge-grounded-framework-for-pedagogical-from.jsonld"}}