This paper introduces a framework to audit source-dependence in multi-source RAG systems, demonstrating that disagreement across institutional sources is a common and critical failure mode that curren…

The paper introduces MIRA, a bilingual benchmark that reveals that LLMs tend to dilute or omit critical medical information when responding to prompts from users with low health literacy, a pattern te…

The paper proposes M extsuperscript{3}Att, a knowledge-poisoning framework that injects covert misinformation into medical multimodal RAG systems using paired visual data triggers, demonstrating attac…

The paper demonstrates that while distilling large language models for medical QA can significantly improve final answer accuracy, this gain often comes at the cost of factual accuracy and detailed re…

The paper introduces DrugClaw, a multi-agent system, and DrugAudit, a new benchmark, demonstrating that DrugClaw excels at answering drug-related questions by grounding answers in primary regulatory s…

The paper proposes a novel, efficient method for checking the factuality of claims generated by LLMs by framing it as a true/false reading comprehension task and incorporating explicit test-taking str…

The paper proposes 'Think Fast, Talk Smart,' a pipeline that separates deterministic data analysis from LLM generation, showing that offloading recurring, structured tasks to code significantly improv…

C-MIG is a novel retrieval-augmented generation framework that uses multi-view information gain to improve clinical diagnosis reasoning by providing richer, more nuanced reward signals than existing m…

The paper introduces CERA, a novel contrastive retrieval framework that improves RAG factuality and interpretability by using subjectivity-based hard negative selection and an auxiliary attention alig…

The paper systematically evaluates advanced retrieval-augmented generation (RAG) architectures for Cyber Threat Intelligence (CTI), demonstrating that a hybrid graph-text approach significantly improv…

The paper introduces CITETRACE, a large-scale dataset and evaluation framework that systematically measures structural citation failures in search-augmented LLMs, revealing a pattern called Verified M…

The paper introduces MedCase-Structured, a synthetic, FHIR-formatted dataset designed to benchmark diagnostic reasoning in realistic EHR settings, showing that LLMs perform worse on structured data th…

The paper introduces Self-Conditioned Positional HNSW (SCP-HNSW), a method that modifies chunk embeddings and retrieval process to mitigate redundant evidence retrieval from overlapping document chunk…

E-MIA introduces a novel, stealthy black-box membership inference attack that converts verifiable hard evidence within a candidate document into an objective, multi-part exam score to determine if the…

The paper introduces MEntA, a highly query-efficient and surrogate-free membership inference attack that uses natural-language entailment to detect if a specific document was used by a RAG system, ach…

The paper introduces FORCEBENCH, a new stress test designed to evaluate whether cited sources genuinely warrant the strength of a claim, revealing that standard citation evaluation methods often fail…

The paper systematically compares multiple content representations for RAG pipelines and finds that answer retention—the ability of the representation to preserve the original answer-bearing content—i…

The paper argues that current search agents often verify existing knowledge rather than genuinely searching, and introduces LiveBrowseComp, a new benchmark to measure true evidence-driven discovery.

This paper demonstrates that patient-facing RAG chatbots frequently expose sensitive system configurations, knowledge base details, and conversation history through client-server communication, posing…

The paper introduces AMNESIA, the first large-scale, open-source benchmark for medical unlearning, demonstrating that current unlearning methods struggle to separate individual patient data from share…