Even when instructed to adhere to source material, Language Models often generate unsubstantiated content - a phenomenon known as "closed-domain hallucination." This risk is amplified in processes with multiple generative steps (MGS), compared to processes with a single generative step (SGS). However, due to the greater complexity of MGS processes, we argue that detecting hallucinations in their final outputs is necessary but not sufficient: it is equally important to trace where hallucinated content was likely introduced and how faithful content may have been derived from the source through intermediate outputs. To address this need, we present VeriTrail, the first closed-domain hallucination detection method designed to provide traceability for both MGS and SGS processes. We also introduce the first datasets to include all intermediate outputs as well as human annotations of final outputs' faithfulness for their respective MGS processes. We demonstrate that VeriTrail outperforms baseline methods on both datasets.

Safety reasoning is a recent paradigm where LLMs reason over safety policies before generating responses, thereby mitigating limitations in existing safety measures such as over-refusal and jailbreak vulnerabilities. However, implementing this paradigm is challenging due to the resource-intensive process of creating high-quality policy-embedded chain-of-thought (CoT) datasets while ensuring reasoning remains accurate and free from hallucinations or policy conflicts. To tackle this, we propose AIDSAFE: Agentic Iterative Deliberation for Safety Reasoning, a novel data generation recipe that leverages multi-agent deliberation to iteratively expand reasoning on safety policies. A data refiner stage in AIDSAFE ensures high-quality outputs by eliminating repetitive, redundant, and deceptive thoughts. AIDSAFE-generated CoTs provide a strong foundation for supervised fine-tuning (SFT)-based safety training. Additionally, to address the need of preference data in alignment stages, such as DPO training, we introduce a supplemental recipe that uses belief augmentation to create distinct selected and rejected CoT samples. Our evaluations demonstrate that AIDSAFE-generated CoTs achieve superior policy adherence and reasoning quality. Consequently, we show that fine-tuning open-source LLMs on these CoTs can significantly improve safety generalization and jailbreak robustness while maintaining acceptable utility and over-refusal accuracy. AIDSAFE-generated CoT datasets can be found here: https://huggingface.co/datasets/AmazonScience/AIDSAFE

Argonne National Laboratory ABSTRACT Nuclear waste management requires rigorous regulatory compliance assessment, demanding advanced decision - support systems capable of addressing complex legal, environmental, and safety considerations. This paper presents a multi - agent Retrieval - Augmented Generation (RAG) system that integrates large language models (LLMs) with document retrieval mechanisms to enhance decision accuracy through structured agent collaboration. Through a structured 10 - round discussion model, agents collaborate to assess regulatory compliance and safety requirements while maintaining document - grounded responses. A case study of a proposed temporary nuclear waste storage site near Winslow, Arizona, demonstrates the framework ' s effectiveness. Results show the Regulatory Agent achieves consistently higher relevance scores in maintaining alignment with legal frameworks, while the Safety Agent effectively manages complex risk assessments requi ring multifaceted analysis. The system demonstrates progressive improvement in agreement rates between agents across discussion rounds while semantic drift decreases, indicating enhanced decision - making consistency and response coherence. The system ensure s regulatory decisions remain factually grounded, dynamically adapting to evolving regulatory frameworks through real - time document retrieval. By balancing automated assessment with human oversight, this framework offers a scalable and transparent approach to regulatory governance. Future research will explore multi - modal data integration and reinforcement learning to enhance response coherence and decision efficiency. These findings underscore the potential of AI - driven, multi - agent systems in advancing ev idence - based, accountable, and adaptive decision - making for high - stakes environmental management scenarios.

As generative AI systems become widely adopted, they enable unprecedented creation levels of synthetic data across text, images, audio, and video modalities. While research has addressed the energy consumption of model training and inference, a critical sustainability challenge remains understudied: digital waste. This term refers to stored data that consumes resources without serving a specific (and/or immediate) purpose. This paper presents this terminology in the AI context and introduces digital waste as an ethical imperative within (generative) AI development, positioning environmental sustainability as core for responsible innovation. Drawing from established digital resource management approaches, we examine how other disciplines manage digital waste and identify transferable approaches for the AI community. We propose specific recommendations encompassing re-search directions, technical interventions, and cultural shifts to mitigate the environmental consequences of in-definite data storage. By expanding AI ethics beyond immediate concerns like bias and privacy to include inter-generational environmental justice, this work contributes to a more comprehensive ethical framework that considers the complete lifecycle impact of generative AI systems.

The persistent accumulation of unresolved legal cases, especially within the Indian judiciary, significantly hampers the timely delivery of justice. Manual methods of prioritizing petitions are often prone to inefficiencies and subjective biases further exacerbating delays. To address this issue, we propose LLMPR (Large Language Model-based Petition Ranking), an automated framework that utilizes transfer learning and machine learning to assign priority rankings to legal petitions based on their contextual urgency. Leveraging the ILDC dataset comprising 7,593 annotated petitions, we process unstructured legal text and extract features through various embedding techniques, including DistilBERT, LegalBERT, and MiniLM. These textual embeddings are combined with quantitative indicators such as gap days, rank scores, and word counts to train multiple machine learning models, including Random Forest, Decision Tree, XGBoost, LightGBM, and CatBoost. Our experiments demonstrate that Random Forest and Decision Tree models yield superior performance, with accuracy exceeding 99% and a Spearman rank correlation of 0.99. Notably, models using only numerical features achieve nearly optimal ranking results (R2 = 0.988, \r{ho} = 0.998), while LLM-based embeddings offer only marginal gains. These findings suggest that automated petition ranking can effectively streamline judicial workflows, reduce case backlog, and improve fairness in legal prioritization.

Artificial Knowledge (AK) systems are transforming decision-making across critical domains such as healthcare, finance, and criminal justice. However, their growing opacity presents governance challenges that current regulatory approaches, focused predominantly on explainability, fail to address adequately. This article argues for a shift toward validation as a central regulatory pillar. Validation, ensuring the reliability, consistency, and robustness of AI outputs, offers a more practical, scalable, and risk-sensitive alternative to explainability, particularly in high-stakes contexts where interpretability may be technically or economically unfeasible. We introduce a typology based on two axes, validity and explainability, classifying AK systems into four categories and exposing the trade-offs between interpretability and output reliability. Drawing on comparative analysis of regulatory approaches in the EU, US, UK, and China, we show how validation can enhance societal trust, fairness, and safety even where explainability is limited. We propose a forward-looking policy framework centered on pre- and post-deployment validation, third-party auditing, harmonized standards, and liability incentives. This framework balances innovation with accountability and provides a governance roadmap for responsibly integrating opaque, high-performing AK systems into society.

While deep neural network (DNN)-based video denoising has demonstrated significant performance, deploying state-of-the-art models on edge devices remains challenging due to stringent real-time and energy efficiency requirements. Computing-in-Memory (CIM) chips offer a promising solution by integrating computation within memory cells, enabling rapid matrix-vector multiplication (MVM). However, existing DNN models are often designed without considering CIM architectural constraints, thus limiting their acceleration potential during inference. To address this, we propose a hardware-algorithm co-design framework incorporating two innovations: (1) a CIM-Aware Architecture, CIM-NET, optimized for large receptive field operation and CIM's crossbar-based MVM acceleration; and (2) a pseudo-convolutional operator, CIM-CONV, used within CIM-NET to integrate slide-based processing with fully connected transformations for high-quality feature extraction and reconstruction. This framework significantly reduces the number of MVM operations, improving inference speed on CIM chips while maintaining competitive performance. Experimental results indicate that, compared to the conventional lightweight model FastDVDnet, CIM-NET substantially reduces MVM operations with a slight decrease in denoising performance. With a stride value of 8, CIM-NET reduces MVM operations to 1/77th of the original, while maintaining competitive PSNR (35.11 dB vs. 35.56 dB

As the impact of large language models increases, understanding the moral values they reflect becomes ever more important. Assessing the nature of moral values as understood by these models via direct prompting is challenging due to potential leakage of human norms into model training data, and their sensitivity to prompt formulation. Instead, we propose to use word associations, which have been shown to reflect moral reasoning in humans, as low-level underlying representations to obtain a more robust picture of LLMs' moral reasoning. We study moral differences in associations from western English-speaking communities and LLMs trained predominantly on English data. First, we create a large dataset of LLM-generated word associations, resembling an existing data set of human word associations. Next, we propose a novel method to propagate moral values based on seed words derived from Moral Foundation Theory through the human and LLM-generated association graphs. Finally, we compare the resulting moral conceptualizations, highlighting detailed but systematic differences between moral values emerging from English speakers and LLM associations.

Large language models are known to memorize parts of their training data, posing risk of copyright violations. To systematically examine this risk, we pretrain language models (1B/3B/8B) from scratch on 83B tokens, mixing web-scale data with public domain books used to simulate copyrighted content at controlled frequencies at lengths at least ten times longer than prior work. We thereby identified the offset effect, a phenomenon characterized by two key findings: (1) verbatim memorization is most strongly triggered by short prefixes drawn from the beginning of the context window, with memorization decreasing counterintuitively as prefix length increases; and (2) a sharp decline in verbatim recall when prefix begins offset from the initial tokens of the context window. We attribute this to positional fragility: models rely disproportionately on the earliest tokens in their context window as retrieval anchors, making them sensitive to even slight shifts. We further observe that when the model fails to retrieve memorized content, it often produces degenerated text. Leveraging these findings, we show that shifting sensitive data deeper into the context window suppresses both extractable memorization and degeneration. Our results suggest that positional offset is a critical and previously overlooked axis for evaluating memorization risks, since prior work implicitly assumed uniformity by probing only from the beginning of training sequences.

Loss reweighting has shown significant benefits for machine unlearning with large language models (LLMs). However, their exact functionalities are left unclear and the optimal strategy remains an open question, thus impeding the understanding and improvement of existing methodologies. In this paper, we identify two distinct goals of loss reweighting, namely, Saturation and Importance -- the former indicates that those insufficiently optimized data should be emphasized, while the latter stresses some critical data that are most influential for loss minimization. To study their usefulness, we design specific reweighting strategies for each goal and evaluate their respective effects on unlearning. We conduct extensive empirical analyses on well-established benchmarks, and summarize some important observations as follows: (i) Saturation enhances efficacy more than importance-based reweighting, and their combination can yield additional improvements. (ii) Saturation typically allocates lower weights to data with lower likelihoods, whereas importance-based reweighting does the opposite. (iii) The efficacy of unlearning is also largely influenced by the smoothness and granularity of the weight distributions. Based on these findings, we propose SatImp, a simple reweighting method that combines the advantages of both saturation and importance. Empirical results on extensive datasets validate the efficacy of our method, potentially bridging existing research gaps and indicating directions for future research. Our code is available at https://github.com/tmlr-group/SatImp.