A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Yagmur Yigit, Mohamed Amine Ferrag, Mohamed Chahine Ghanem, et al.
A review of AI-driven approaches for critical infrastructure protection, covering LLM cybersecurity benchmarks, trust and privacy issues, and agentic AI for proactive defense.
Unknown
HT2ML combines homomorphic encryption and trusted execution environments to enable efficient privacy-preserving machine learning on outsourced cloud data.
Unknown
FaithfulRAG introduces fact-level conflict modeling to improve context faithfulness in retrieval-augmented generation, with potential extensions to multimodal RAG and trustworthy AI.
Unknown
A survey examining trustworthiness challenges in retrieval-augmented generation for large language models, covering evaluation, robustness, and open problems.
Unknown
An empirical study showing LLM agents can both help and harm user trust and team performance when acting as daily assistants, depending on how plans are executed.
J. Jakeman, Lorena A. Barba, Joaquim R. R. A. Martins, et al.
This paper proposes a framework for verification and validation (V&V) of scientific machine learning models to enhance trustworthiness and reliability in scientific applications.
Unknown
This paper proposes a framework to harden LLM fine-tuning by combining differentially private data selection with trustworthiness-aware model quantization to enhance privacy and robustness.
Unknown
This paper introduces Multitrust, a comprehensive benchmark for evaluating the trustworthiness of multimodal large language models across multiple dimensions.
Xian Sun, Wei Chow, Yingshuo Wang, et al.
Introduces MIST benchmark and SC2W metric for selective trust in LLMs, plus SCOPE method using balanced DPO to reduce context-induced errors without sacrificing accuracy.
Unknown
This paper investigates how trust in AI creators influences trust in AI systems, emphasizing the critical roles of AI alignment and steerability within the agentic IS framework.
Aniri, Jinhe Bi, Peng Liao, et al.
OPD-V introduces a visual on-policy self-distillation paradigm that uses modality-balance trust regions to improve MLLM reasoning while reducing training cost.
Zohaib Salahuddin, Henry C. Woodruff, Avishek Chatterjee, et al.
This review categorizes nine interpretability methods for deep neural networks in medical image analysis, evaluates progress in explanation assessment, and provides guidelines and future directions for trustworthy AI.