LLM-Generated Text May Harm Your Retrieval! A Robust Detection Strategy for Retrieval-Augmented Generation

Integrating Retrieval-Augmented Generation with Large Language Model for Robust and Explainable AI Text Detection

Large Language Models (LLMs) have been rapidly evolving lately, resulting in the need for strong, explainable models to detect the difference between human-generated and machine-generated articles. Existing approaches which are mostly based on fine-tuned transformers suffer from several drawbacks such as rapid obsolescence, paraphrasing attacks, and lack of interpretability. To improve their ability to detect, this paper proposes a novel paradigm called Human vs. LLM Identification (HLI) which introduces a Retrieval-Augmented Generation (RAG)-inspired evidence-based detection strategy alongside a fine-tuned transformer classifier. Our core model, DeBERTa-Sentinel, is built on top of a fine-tuned Microsoft DeBERTa-v3-small model, which uses a disentangled attention mechanism to better capture subtle syntactic and stylistic deviations characteristic of AI-generated text. We evaluate our framework on a balanced dataset of 43,456 text samples, curated from the OpenGPTText corpus and covering AI-generated and human-authored content across diverse domains including news, education, and creative text. The experimental results show improved performance over the selected baselines, with our framework achieving an accuracy of 97.53%, precision of 95.89%, recall of 99.34%, and ROC-AUC of 99.53%. In addition, explainability is integrated into our framework through Local Interpretable Model-agnostic Explanations (LIME) analysis, providing token-level insight into classification decisions. This study establishes a benchmark for scalable, explainable AI text detection, with implications for academic integrity, content moderation, and combating misinformation.

Exploring Internal Emphasis for Robust Noisy RAG

Benchmarking retrieval augmented generation LLMs for Arabic noise robustness

Hallucination has become a serious concern in large language models (LLMs), as these models can generate useful yet incorrect or misleading information, which has led to growing research interest in retrieval-augmented generation (RAG) as a mitigation approach. RAG provides LLMs with access to external information, such as databases or documents, which can help them to answer users' questions. Currently, several benchmarks released measure RAG performance on various LLMs; however, evaluations of the noise robustness ability in Arabic are absent. In this paper, we systematically investigated the capabilities of state-of-the-art multilingual LLMs with regard to two essential RAG abilities, noise robustness and negative rejection. To accomplish this, we generated an Arabic benchmark consisting of 300 questions along with 6,196 documents. Then, we assessed the performance of six LLMs in relation to the two aforementioned RAG abilities. The results reveal that all six LLMs were negatively affected when the noise ratio in the external documents was increased. Under the highest noise settings at 80%, the best LLM performance was for Claude-4 sonnet, in which their performance decreased by only 4.67 percentage points. Furthermore, when it comes to the negative rejection task, there has been a significant impact on all six models. The best two models, Claude-4 sonnet and Llama-4, scored 90.67% and 85.67%, respectively, while smaller models, like GPT-3.5, scored 69.33%. Furthermore, our manual analysis reveals that many errors made by LLMs are attributed to over-caution behavior. LLMs often decline to respond probably due to training mechanisms designed to reduce hallucinations. Additionally, other errors occur when there is a high lexical similarity between the question and the words of noisy documents, which causes the model to rely on irrelevant content instead of the correct information.

Feedback-Guided Prompt Injection Defense in Retrieval-Augmented Text-to-Cypher Generation

A dynamic and self-corrective procedure with feedback-loop-based AI architecture with Large Language Models (LLMs) for near real-time validation and correction of generated queries and it is demonstrated that the proposed solution outperforms various training-free prompt injection defense methods.

SAC-RAG: Semantic Adaptive Context Compression for Retrieval-Augmented Generation

Experimental results show that SAC-RAG reduces token consumption by 38%–58% at the cost of only a 1–2 percentage point EM drop, with EM actually improving after compression for reasoning-type questions, achieving the optimal quality–efficiency trade-off in terms of token consumption.

Robust and explainable Retrieval-Augmented Generation under retrieval noise

Retrieval-Augmented Generation (RAG) improves language-model answers by retrieving external evidence before generation. However, its reliability depends on the retrieved context. In real settings, passages can be irrelevant, incomplete, conflicting, or poorly ordered. These problems may reduce accuracy and explainability. This study tests how retrieval noise affects RAG and whether reranking, citation-aware generation, and lightweight verification can improve system behaviour. A controlled experiment was conducted on a small HotpotQA subset using BM25, Sentence-BERT, and FAISS. Four systems were compared: vanilla RAG, reranking-only, citation-only, and a full enhanced system. Results show that reranking achieved the highest average noisy F1, but the gain over vanilla RAG was small. The full enhanced system achieved better faithfulness, groundedness, and citation precision, but did not improve average noisy F1. This suggests that robust and explainable RAG is a multi-objective problem.