LayerRAG-Bench: A Cross-Layer Reliability Benchmark for Agentic Retrieval-Augmented Generation

arXiv:2607.27353v1 Announce Type: new Abstract: Agentic retrieval-augmented generation systems can produce answers that appear grounded while failing at the evidence, tool-contract, authorization, or session-state layer. We introduce LayerRAG-Bench, a controlled cross-layer reliability benchmark with 8 enterprise domains, 240 tasks, 9 fault scenarios, 2 contract modes, and 38,880 live task-level records across nine models from OpenAI, Anthropic, and Gemini. Schema normalization raises schema-drift success from 0.000 to 0.913, but stale evidence, missing tool output, denied permissions, and wrong-session context are not recovered by schema normalization. Groundedness-only evaluation also produces substantial false positives under stale and wrong-session evidence. These results support a layer-specific evaluation principle: a reliability intervention should be credited for repairing its target layer without being mistaken for a universal fix.

Related

• RAGCap-Bench: Benchmarking Capabilities of LLMs in Agentic Retrieval Augmented Generation Systems
• SURE-RAG: Sufficiency and Uncertainty-Aware Evidence Verification for Selective Retrieval-Augmented Generation
• Beyond Benchmark Islands: Toward Representative Trustworthiness Evaluation for Agentic AI
• Rethinking Reasoning-Intensive Retrieval: Evaluating and Advancing Retrievers in Agentic Search Systems

Source: arXiv cs.CL | 2026-07-31