arXiv:2510.24120v1 Announce Type: new 
Abstract: Graph-based RAG constructs a knowledge graph (KG) from text chunks to enhance retrieval in Large Language Model (LLM)-based question answering. It is especially beneficial in domains such as biomedicine, law, and political science, where effective retrieval often involves multi-hop reasoning over proprietary documents. However, these methods demand numerous LLM calls to extract entities and relations from text chunks, incurring prohibitive costs at scale. Through a carefully designed ablation study, we observe that certain words (termed concepts) and their associated documents are more important. Based on this insight, we propose Graph-Guided Concept Selection (G2ConS). Its core comprises a chunk selection method and an LLM-independent concept graph. The former selects salient document chunks to reduce KG construction costs; the latter closes knowledge gaps introduced by chunk selection at zero cost. Evaluations on multiple real-world datasets show that G2ConS outperforms all baselines in construction cost, retrieval effectiveness, and answering quality.

تُحدث طريقة جديدة تُدعى RAG المعتمدة على الرسوم البيانية ضجة في مجال الإجابة على الأسئلة من خلال بناء رسم بياني للمعرفة من قطع النص. تُحسن هذه الطريقة بشكل كبير من كفاءة الاسترجاع، خاصة في المجالات المعقدة مثل الطب الحيوي والقانون والعلوم السياسية، حيث يكون التفكير متعدد الخطوات أمرًا حاسمًا. من خلال تبسيط عملية استخراج الكيانات والعلاقات، تعد بتقليل التكاليف المرتبطة باستخدام نماذج اللغة الكبيرة، مما يجعل تقنيات الاسترجاع المتقدمة أكثر سهولة وفعالية.

Un nuevo enfoque llamado RAG basado en grafos está causando sensación en el campo de la respuesta a preguntas al construir un grafo de conocimiento a partir de fragmentos de texto. Este método mejora significativamente la eficiencia de la recuperación, especialmente en dominios complejos como la biomedicina, el derecho y las ciencias políticas, donde el razonamiento de múltiples saltos es crucial. Al optimizar el proceso de extracción de entidades y relaciones, promete reducir los costos asociados con el uso de grandes modelos de lenguaje, haciendo que las técnicas de recuperación avanzadas sean más accesibles y efectivas.

Une nouvelle approche appelée RAG basé sur des graphes fait sensation dans le domaine de la réponse aux questions en construisant un graphe de connaissances à partir de morceaux de texte. Cette méthode améliore considérablement l'efficacité de la récupération, en particulier dans des domaines complexes comme la biomédecine, le droit et les sciences politiques, où le raisonnement multi-saut est crucial. En rationalisant le processus d'extraction des entités et des relations, elle promet de réduire les coûts associés à l'utilisation de grands modèles de langage, rendant les techniques de récupération avancées plus accessibles et efficaces.

A new approach called Graph-based RAG is making waves in the field of question answering by constructing a knowledge graph from text chunks. This method significantly enhances retrieval efficiency, particularly in complex domains like biomedicine, law, and political science, where multi-hop reasoning is crucial. By streamlining the process of extracting entities and relations, it promises to reduce the costs associated with using large language models, making advanced retrieval techniques more accessible and effective.

Graph-Guided Concept Selection for Efficient Retrieval-Augmented Generation

arXiv:2601.08209v1 Announce Type: new 
Abstract: In domains such as biomedicine, materials, and finance, high-stakes deployment of large language models (LLMs) requires injecting private, domain-specific knowledge that is proprietary, fast-evolving, and under-represented in public pretraining. However, the two dominant paradigms for private knowledge injection each have pronounced drawbacks: fine-tuning is expensive to iterate, and continual updates risk catastrophic forgetting and general-capability regression; retrieval-augmented generation (RAG) keeps the base model intact but is brittle in specialized private corpora due to chunk-induced evidence fragmentation, retrieval drift, and long-context pressure that yields query-dependent prompt inflation. Inspired by how multimodal LLMs align heterogeneous modalities into a shared semantic space, we propose Generation-Augmented Generation (GAG), which treats private expertise as an additional expert modality and injects it via a compact, representation-level interface aligned to the frozen base model, avoiding prompt-time evidence serialization while enabling plug-and-play specialization and scalable multi-domain composition with reliable selective activation. Across two private scientific QA benchmarks (immunology adjuvant and catalytic materials) and mixed-domain evaluations, GAG improves specialist performance over strong RAG baselines by 15.34% and 14.86% on the two benchmarks, respectively, while maintaining performance on six open general benchmarks and enabling near-oracle selective activation for scalable multi-domain deployment.

تم اقتراح إطار جديد يسمى Generation-Augmented Generation (GAG) لتحسين حقن المعرفة الخاصة والمحددة في مجالات معينة في نماذج اللغة الكبيرة (LLMs)، مما يعالج التحديات في مجالات مثل الطب الحيوي والمواد والتمويل. تهدف هذه الطريقة إلى التغلب على قيود الضبط الدقيق والتوليد المعزز بالاسترجاع من خلال اعتبار الخبرة الخاصة كنوع إضافي من الخبرة.

Se ha propuesto un nuevo marco llamado Generation-Augmented Generation (GAG) para mejorar la inyección de conocimientos privados y específicos de dominio en grandes modelos de lenguaje (LLMs), abordando desafíos en campos como la biomedicina, los materiales y las finanzas. Este enfoque busca superar las limitaciones del ajuste fino y la generación aumentada por recuperación al tratar la experiencia privada como una modalidad experta adicional.

Un nouveau cadre appelé Generation-Augmented Generation (GAG) a été proposé pour améliorer l'injection de connaissances privées et spécifiques à un domaine dans les grands modèles de langage (LLMs), répondant aux défis dans des domaines tels que la biomédecine, les matériaux et la finance. Cette approche vise à surmonter les limitations du fine-tuning et de la génération augmentée par récupération en considérant l'expertise privée comme une modalité experte supplémentaire.

A new framework called Generation-Augmented Generation (GAG) has been proposed to enhance the injection of private, domain-specific knowledge into large language models (LLMs), addressing challenges in fields like biomedicine, materials, and finance. This approach aims to overcome the limitations of fine-tuning and retrieval-augmented generation by treating private expertise as an additional expert modality.

Generation-Augmented Generation: A Plug-and-Play Framework for Private Knowledge Injection in Large Language Models

arXiv:2510.21060v2 Announce Type: replace 
Abstract: Policy optimization (PO) is a cornerstone of modern reinforcement learning (RL), with diverse applications spanning robotics, healthcare, and large language model training. The increasing deployment of PO in sensitive domains, however, raises significant privacy concerns. In this paper, we initiate a theoretical study of differentially private policy optimization, focusing explicitly on its sample complexity. We first formalize an appropriate definition of differential privacy (DP) tailored to PO, addressing the inherent challenges arising from on-policy learning dynamics and the subtlety involved in defining the unit of privacy. We then systematically analyze the sample complexity of widely-used PO algorithms, including policy gradient (PG), natural policy gradient (NPG) and more, under DP constraints and various settings, via a unified framework. Our theoretical results demonstrate that privacy costs can often manifest as lower-order terms in the sample complexity, while also highlighting subtle yet important observations in private PO settings. These offer valuable practical insights for privacy-preserving PO algorithms.

تم نشر دراسة حديثة حول تحسين السياسات بشكل مختلف (DPPO) تركز على تعقيد العينة في تحسين السياسات (PO) في التعلم المعزز (RL). تتناول هذه الدراسة المخاوف المتعلقة بالخصوصية في التطبيقات الحساسة مثل الروبوتات والرعاية الصحية من خلال صياغة تعريف للخصوصية المختلفة يتناسب مع PO وتحليل تعقيد العينة لعدة خوارزميات PO تحت قيود DP.

Se ha publicado un estudio reciente sobre la optimización de políticas diferencialmente privadas (DPPO), centrado en la complejidad de muestra de la optimización de políticas (PO) en el aprendizaje por refuerzo (RL). Esta investigación aborda las preocupaciones de privacidad en aplicaciones sensibles como la robótica y la atención médica al formalizar una definición de privacidad diferencial adaptada a la PO y analizar la complejidad de muestra de varios algoritmos de PO bajo restricciones de DP.

Une étude récente sur l'optimisation de politiques différemment privées (DPPO) a été publiée, se concentrant sur la complexité d'échantillonnage de l'optimisation de politiques (PO) dans l'apprentissage par renforcement (RL). Cette recherche aborde les préoccupations en matière de confidentialité dans des applications sensibles telles que la robotique et les soins de santé en formalisant une définition de la confidentialité différentielle adaptée à la PO et en analysant la complexité d'échantillonnage de divers algorithmes de PO sous des contraintes de DP.

A recent study on differentially private policy optimization (DPPO) has been published, focusing on the sample complexity of policy optimization (PO) in reinforcement learning (RL). This research addresses privacy concerns in sensitive applications such as robotics and healthcare by formalizing a definition of differential privacy tailored to PO and analyzing the sample complexity of various PO algorithms under DP constraints.

On the Sample Complexity of Differentially Private Policy Optimization

One More Thing in AI – Your Shortcut to AI Mastery

Graph-Guided Concept Selection for Efficient Retrieval-Augmented Generation

Was this article worth reading? Share it

One More Thing in AI

LucidQuery AI

Magicley AI

Airparser

Supametas.AI

Graphite Note

Ready to build your own newsroom?