# DynaContext: Self-Improving Dynamic Contextualization of Optimized Prompts for Heterogeneous Parameter Extraction

arXiv: https://arxiv.org/abs/2608.22014 · HTML: https://mlsanity.com/paper/2608.22014/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Self-rewarding alignment and self-training, one of 13 directions in Synthetic data and self-improvement for LLMs.
- Builds on 25 of its 38 references that are in the map.
- First posted on arXiv 22 Aug 2026.

## Placed in

- Synthetic data and self-improvement for LLMs / [Self-rewarding alignment and self-training](https://mlsanity.com/synth/self-rewarding-alignment-and-self-training/)

## Builds on (most cited first, 25 of 25)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines](https://mlsanity.com/paper/2310.03714/) (2023)
- [GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning](https://mlsanity.com/paper/2507.19457/) (2025)
- [Optimizing Instructions and Demonstrations for Multi-Stage Language Model Programs](https://mlsanity.com/paper/2406.11695/) (2024)
- [OffsetBias: Leveraging Debiased Data for Tuning Evaluators](https://mlsanity.com/paper/2407.06551/) (2024)
- [OpenTag: Open Attribute Value Extraction from Product Profiles [Deep Learning, Active Learning, Named Entity Recognition]](https://mlsanity.com/paper/1806.01264/) (2018)
- [MAVE: A Product Dataset for Multi-source Attribute Value Extraction](https://mlsanity.com/paper/2112.08663/) (2021)
- [You Don't Need Pre-built Graphs for RAG: Retrieval Augmented Generation with Adaptive Reasoning Structures](https://mlsanity.com/paper/2508.06105/) (2025)
- [Think-J: Learning to Think for Generative LLM-as-a-Judge](https://mlsanity.com/paper/2505.14268/) (2025)
- [ExtractGPT: Exploring the Potential of Large Language Models for Product Attribute Value Extraction](https://mlsanity.com/paper/2310.12537/) (2023)
- [Using LLMs for the Extraction and Normalization of Product Attribute Values](https://mlsanity.com/paper/2403.02130/) (2024)
- [Enhancing Chain of Thought Prompting in Large Language Models via Reasoning Patterns](https://mlsanity.com/paper/2404.14812/) (2024)
- [TrustUQA: A Trustful Framework for Unified Structured Data Question Answering](https://mlsanity.com/paper/2406.18916/) (2024)
- [ReCode: Updating Code API Knowledge with Reinforcement Learning](https://mlsanity.com/paper/2506.20495/) (2025)
- [Multi-Label Zero-Shot Product Attribute-Value Extraction](https://mlsanity.com/paper/2402.08802/) (2024)
- [Toward Reliable Evaluation of LLM-Based Financial Multi-Agent Systems: Taxonomy, Coordination Primacy, and Cost Awareness](https://mlsanity.com/paper/2603.27539/) (2026)
- [Retrieval Augmented Structured Generation: Business Document Information Extraction As Tool Use](https://mlsanity.com/paper/2405.20245/) (2024)
- [Schema as Parameterized Tools for Universal Information Extraction](https://mlsanity.com/paper/2506.01276/) (2025)
- [Agent-UniRAG: A Trainable Open-Source LLM Agent Framework for Unified Retrieval-Augmented Generation Systems](https://mlsanity.com/paper/2505.22571/) (2025)
- [GRETEL: A Goal-driven Retrieval and Execution-based Trial Framework for LLM Tool Selection Enhancing](https://mlsanity.com/paper/2510.17843/) (2025)
- [Self-Correction Distillation for Structured Data Question Answering](https://mlsanity.com/paper/2511.07998/) (2025)
- [Retrieval-Augmented Multi-LLM Ensemble for Industrial Part Specification Extraction](https://mlsanity.com/paper/2601.05266/) (2025)
- [Teaching Models to Improve on Tape](https://mlsanity.com/paper/2411.01483/) (2024)
- [Enhancing Relation Extraction via Supervised Rationale Verification and Feedback](https://mlsanity.com/paper/2412.07289/) (2024)
- [Dynamic Deep Prompt Optimization for Defending Against Jailbreak Attacks on LLMs](https://mlsanity.com/paper/2609.26185/) (2026)

## Abstract (arXiv)

Automated prompt and skill optimization typically produces a single static instruction that is reused across inference instances until the next optimization cycle. However, this approach cannot adapt when the required context, constraints, and evidence vary from one instance to another. For instance, parameter extraction from electronic component descriptions breaks this assumption: resistors, capacitors, transistors, and connectors require different fields, unit constraints, and demonstrations, and each input provides a different evidence state. We introduce DynaContext, a framework that combines an offline-optimized extraction core, learned with GEPA or SkillOpt, with inference-time contextual adaptation and validation-gated self-improvement. DynaContext routes each item through internal, external, or fallback evidence paths and composes an item-specific prompt from the core, schema, evidence, unresolved fields, and validated demonstrations. Deterministic validation and an LLM judge gate every output, uncertain cases go to human review, and only human-verified corrections enter the demonstration memory. On a single-category benchmark, average accuracy increases from 86.6% for the base prompt to 96.9% for standalone SkillOpt and 98.6% for the best DynaContext configuration. Across 850 heterogeneous gold parameter facts, average field-level F1 increases from 51.8% for an unoptimized, demonstration-free control to 59.2% with dynamic demonstrations alone, 66.9% with the optimized core alone, and 71.0% with both. Holding the model fixed, the full configuration outperforms the deployed static-prompting pipeline by 17.3 F1 points on average.