# No=esis: Deterministic-First Retrieval with Two-Tier Context Hydration for Factuality-Critical Queries on Small Local Models

arXiv: https://arxiv.org/abs/2609.07663 · HTML: https://mlsanity.com/paper/2609.07663/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 30 references that are in the map.
- First posted on arXiv 7 Sep 2026.

## Builds on (most cited first, 22 of 22)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection](https://mlsanity.com/paper/2310.11511/) (2023)
- [From Local to Global: A Graph RAG Approach to Query-Focused Summarization](https://mlsanity.com/paper/2404.16130/) (2024)
- [MemGPT: Towards LLMs as Operating Systems](https://mlsanity.com/paper/2310.08560/) (2023)
- [RECOMP: Improving Retrieval-Augmented LMs with Compression and Selective Augmentation](https://mlsanity.com/paper/2310.04408/) (2023)
- [Corrective Retrieval Augmented Generation](https://mlsanity.com/paper/2401.15884/) (2024)
- [RetroLLM: Empowering Large Language Models to Retrieve Fine-grained Evidence within Generation](https://mlsanity.com/paper/2412.11919/) (2024)
- [COMI: Coarse-to-fine Context Compression via Marginal Information Gain](https://mlsanity.com/paper/2602.01719/) (2026)
- [The Missing Memory Hierarchy: Demand Paging for LLM Context Windows](https://mlsanity.com/paper/2603.09023/) (2026)
- [SymRAG: Efficient Neuro-Symbolic Retrieval Through Adaptive Query Routing](https://mlsanity.com/paper/2506.12981/) (2025)
- [Retrieval as Generation: A Unified Framework with Self-Triggered Information Planning](https://mlsanity.com/paper/2604.11407/) (2026)
- [CLAUSE: Agentic Neuro-Symbolic Knowledge Graph Reasoning via Dynamic Learnable Context Engineering](https://mlsanity.com/paper/2509.21035/) (2025)
- [ProgRAG: Hallucination-Resistant Progressive Retrieval and Reasoning over Knowledge Graphs](https://mlsanity.com/paper/2511.10240/) (2025)
- [Can Small Language Models Use What They Retrieve? An Empirical Study of Retrieval Utilization Across Model Scale](https://mlsanity.com/paper/2603.11513/) (2026)
- [Structure Guided Retrieval-Augmented Generation for Factual Queries](https://mlsanity.com/paper/2604.22843/) (2026)
- [MC-RAG System: A Structure-Driven RAG System for Multi-Constraint Queries](https://mlsanity.com/paper/2607.10151/) (2026)
- [Query Symbolically or Retrieve Semantically? A Dataset and Method for Semi-Structured Question Answering](https://mlsanity.com/paper/2605.27164/) (2026)
- [Models for minimalist RAG: B1ade 335M Embedding and 1B Parameter Small Language Models](https://mlsanity.com/paper/2607.27506/) (2026)
- [SeDeM: Selective Decompression of Hidden-State Memories for Long-Context Question Answering](https://mlsanity.com/paper/2608.00311/) (2026)
- [Noesis: Bidirectional Graph-RAG with Adaptive Parallelism and Cross-Knowledge-Base Semantic Discovery](https://mlsanity.com/paper/2608.15919/) (2026)
- [TopoCompress: Long Context Compression via Graph-Wired Semantic Trajectories](https://mlsanity.com/paper/2608.30811/) (2026)

## Abstract (arXiv)

A wrong number is worse than no answer. Across factuality-critical domains -- audience metrics, scheduling and rights in media; dosages and lab values in healthcare; figures and citations in finance and legal -- a confident but fabricated value is more damaging than an honest admission of uncertainty. Yet this is the dominant failure mode we observe on small local language models: even when correct evidence is present in context, models fabricate plausible numbers and timestamps. Recent work characterizes a real limit of this regime: below 7B parameters, the bottleneck of retrieval-augmented generation (RAG) is not retrieval quality but context utilization. We present Noesis, the deterministic-first query plane of the Noesis architecture, which makes every deterministic judgment before generation. Its mechanisms follow from the ingestion architecture (subject of a separate patent application): (a) a producer-side fact layer rendering precomputed metric facts verbatim without ranking; (b) positional addressing with deterministic cross-source alignment, resolved ahead of query time at zero LLM cost; (c) provenance scoping as an attribution constraint with multi-tier named-reference routing; and (d) two-tier context with model-triggered verbatim hydration. Across four ablations, a 2B model reaches parity with a 35B model on factual integrity (exact values in all runs; zero confabulated numbers on absent-entity traps); structured retrieval beats flat RAG by +11.4 points at 2B; skeleton-only context preserves quantitative answers at 20-30% smaller prompts; and hydration recovers verbatim narrative in ~8s versus ~29s. Two properties matter for regulated domains: each query resolves in a single generation call, and every reported value is traceable to its exact source and position by construction.