Safety & alignment

BackgroundCancer recurrence and distant metastasis are major causes of cancer-related death, yet existing biomarkers and single-omics models have limited accuracy and interpretability across tumor types.MethodsWe developed OMNIS (OMics Network Integration and Spatial representation), a convolutional deep-learning framework that embeds multi-omics profiles into a five-channel genomic image ordered by Hi-C–derived chromosomal proximity. Somatic mutation, copy-number alteration, DNA methylation and

Who has two thumbs and is broken inside?

While recent advances in 3D generation have enabled impressive visual synthesis, existing methods often rely on 2D diffusion supervision without explicit mechanisms for geometric consistency, leading to spatial hallucinations such as duplicated structures and misaligned geometry. These issues become more severe in 4D generation, where maintaining consistency across viewpoints and temporal evolution introduces additional challenges, including jitter, identity flicker, and structural drift. We pre

Multi-reference-to-audio-video (MR2AV) generation aims to generate coherent audio-video content conditioned on multiple references and textual instructions. Existing benchmarks mainly focus on text-driven generation, single-reference subject preservation, or isolated audio-video alignment, leaving the emerging MR2AV setting largely unexplored. Compared with these settings, MR2AV requires models to jointly reason over multiple references while generating synchronized visual and audio content. Mod

Finetuning a pretrained vision-language model (VLM) on robot demonstrations via behavior cloning (BC) has become the standard recipe for vision-language-action (VLA) policies. However, BC finetuning progressively overwrites the pretrained representations that support visual and semantic generalization. Co-training on web image-text data, a common remedy, does not prevent this; it applies language and action losses to separate observations, leaving VLAs with language-action misalignment that stan

Publication date: June 2026 Source: Computers and Education: Artificial Intelligence, Volume 10 Author(s): Tongxi Liu, Luyao Ye, Wei Yan

Publication date: June 2026 Source: Computers and Education: Artificial Intelligence, Volume 10 Author(s): Joyce W. Lacy, Chi Nnoka, Zachary Jock, Cathleen Morreale

Background: Large language models (LLMs) show growing potential for decision support. However, integrating domain-specific medical knowledge while maintaining accuracy, safety, and interpretability remains challenging for postoperative discharge instructions and patient education. Fine-tuning, retrieval-augmented generation (RAG), and hybrid fine-tuning+RAG approaches are prominent strategies for knowledge integration, but their comparative performance in postoperative care has not been systemat

Multi-agent Reinforcement Learning (MARL) holds great potential for robot swarms, but the black-box nature of neural policies complicates strategic analysis, limiting multi-robot applications. Furthermore, complex swarm behaviors can surprisingly emerge from simple rewards without explicit aggregation incentives. Unveiling the mechanisms behind this emergence is critical, but the disconnection between simple rewards and collective behaviors exacerbates interpretability challenges. This paper aim

Jailbreak-robustness research typically evaluates safety through generated responses using an LLM-as-judge approach. Such evaluations, however, are sensitive to the benchmark's grading procedure and capture only observed behavior on a given set of attacks, without directly revealing the hidden fragility of the underlying safety mechanisms. This work proposes JADR (Jacobian Assessment of Danger Recognition), a protocol that measures a model's internal representation through Jacobian space (J-spac

While recent advances in 3D generation have enabled impressive visual synthesis, existing methods often rely on 2D diffusion supervision without explicit mechanisms for geometric consistency, leading to spatial hallucinations such as duplicated structures and misaligned geometry. These issues become more severe in 4D generation, where maintaining consistency across viewpoints and temporal evolution introduces additional challenges, including jitter, identity flicker, and structural drift. We pre

While recent advances in 3D generation have enabled impressive visual synthesis, existing methods often rely on 2D diffusion supervision without explicit mechanisms for geometric consistency, leading to spatial hallucinations such as duplicated structures and misaligned geometry. These issues become more severe in 4D generation, where maintaining consistency across viewpoints and temporal evolution introduces additional challenges, including jitter, identity flicker, and structural drift. We pre

Aligned language models refuse harmful requests, but a one-line prefill ("Sure, here is") strips the refusal. We ask where and how it fails. The harm representation stays intact: on the prompts the attack flips to compliance, a linear probe reads harm as high as on the refused ones (0.91-0.98), while behavioral refusal drops to chance. This holds across four models and three families (1.5-3.8B, and at 14B). Refusal is therefore a shallow, response-site computation. We localize it to an early win

If your institution has deployed a generative AI system in production, you have a guardrail layer. T...

The capability of LLM agents to function as the ``brain'' of a system fundamentally expands the scope of analysis beyond a standalone model. Consequently, safety is no longer only about input--output content alignment. It also concerns system behavior and real-world execution outcomes. However, the current literature is fragmented across attack types, applications, and benchmarks. This makes it hard to explain why failures such as prompt injection, tool misuse, and memory poisoning often share t

Sparse autoencoders (SAEs) are the standard for decomposing superposed neural representations into interpretable features, and evaluation relies predominantly on correlational recovery metrics -- cosine similarity between ground-truth directions and decoder atoms. We show this conflates two distinct claims: decoder-geometry alignment and encoder-activation behavior. We reproduce the superposition phase diagram of Elhage et al. (2022), identifying a convergence artifact at high sparsity and an un

Evaluating retail conversational agents requires methods beyond lexical-overlap metrics to assess intent alignment, factuality, helpfulness, clarity, tone, and overall response quality. Although LLM-as-a-judge methods provide scalable alternatives to human evaluation, production deployment introduces challenges in governance, reproducibility, cost, schema consistency, traceability, and reliability. We present GenAI Evaluation, a governed, configuration-driven pipeline for large-scale evaluation

Online shopping is increasingly shifting toward a model in which AI agents independently search for products, compare options, evaluate constraints, and carry out parts of the purchasing process for users. Website design must now support both human and agent-mediated interaction. This paper introduces the agent-ready website, a design framework for enhancing the readability, interpretability, verifiability, and actionability of e-commerce platforms for AI agents. Existing web design, SEO, and ge

Existing studies of LLM-as-judge scoring bias work predominantly at the input-output level: they perturb inputs, measure score deltas, and propose prompt-level mitigations. We argue that the same biases admit a representation-level account in the judge's hidden state, complementary to the input-output view and operationally useful in ways it does not afford. We report three findings, across seven judges, seven bias types, and nine benchmarks. Geometry: baseline judging inputs occupy a tight acti

Shared meaning in language requires people to learn and agree on categories. We ask how characteristics of agents' memories change the emergence and evolution of shared meaning. Without a coordination game, models of conceptual semantics cannot explain how shared meaning emerges and changes in groups of people; however, existing games assume that players share payoffs in a partnership setting. We model conceptual alignment as a non-partnership game and illustrate differences in actual and percei

Production LLM agents such as Claude Code and Codex operate over untrusted content, files, commands, and workspace state, making safety failures directly actionable. Red-teaming must therefore keep pace with evolving models and tools. Existing approaches mainly optimize attack success and preserve artifacts such as benchmarks, payloads, or attack programs, which record where attacks succeed but not the enabling conditions behind unsafe agent behavior. We study automated red-teaming for productio

Distributional reinforcement learning agents learn full return distributions that are increasingly read at face value: for interpretability, risk-sensitive control, and safety monitoring. We ask a question theory anticipates but that has not been measured directly: are the risk claims of a trained distributional agent true? Our audit combines a decision-relevant screening metric (the excess Wasserstein gap between the top two actions, which equals the mass by which first-order stochastic dominan

Security teams have spent years trying to see more. More endpoints, more cloud services, more identities, more telemetry. For the The post What an ex-NSA red teamer wants every SOC to stop doing appeared first on The New Stack .

China’s Ministry of Industry and Information Technology (MIIT) has started building a safety benchmark to evaluate artificial intelligence models, as regulators in the United States and Europe strengthen oversight of AI security. The MIIT-led National Industrial Information Security Development Research Centre is now recruiting companies and experts to co-build the benchmark, with applications due on Tuesday, according to a notice published on Monday. The institute said that current frameworks..

Post-training is essential for refining the domain-specific capabilities of large language models (LLMs), yet existing reward optimization and distribution matching methods tightly couple policy exploration with distribution alignment. This coupling forces expensive exploration directly on the policy model and severely hinders the asynchronous generation, reuse, and cross-model transfer of optimization signals. In this paper, we propose Proxy-guided Update Signal Transfer (PUST), a novel post-tr

Agentic systems increasingly improve themselves by editing skills: prompts, rubrics, plans, tool contracts, examples, validators, and traces. Skill edits are not independent coordinates in a vector space: they are local repairs to structured artifacts whose effects are observed only after rollout, validation, and critique. Distinct edits can have the same immediate visible effect while differing in routing context, template state, guardrail scope, or future composability. The order of edits can

Safety alignment in large language models can be fragile under fine-tuning, as even benign task adaptation may increase harmful compliance. Existing defenses mainly follow two directions: they either intervene during or after fine-tuning through retraining or weight modification, which can be costly and may hurt task performance, or they use model-agnostic safety classifiers, which may miss failures specific to a given fine-tuned checkpoint. These limitations motivate a post hoc, model-specific,

Cryptocurrency markets are vulnerable to trade-based manipulation, such as wash trading, which can distort price signals and mislead investors. Prior research has mainly focused on detecting manipulation using fixed rules or labeled examples, offering limited flexibility and interpretability for assessing potential risks. Existing visual analytics tools can reveal basic manipulation-related signals, such as token distribution, but still require substantial manual effort to integrate holder relat

Graph Foundation Models (GFMs) have emerged as a promising paradigm for learning transferable representations across diverse graph domains. Recent advancements in GFMs have been largely dominated by two paradigms: Graph Neural Network and Large Language Model (LLM) based methods. However, these methods often face a fundamental dilemma between training with limited data and a heavy reliance on textual attributes. Tabular foundation models (TFMs) offer a potential alternative, as node features and

Generating immersive, synchronized and cinematic audio for long-form textual narratives remains a significant challenge in multi-modal AI. While current Text-to-Audio (TTA) frameworks successfully synthesize isolated sound effects, they struggle with narrative cohesion, temporal alignment, and cinematic emotional depth. We present BackgroundMellow, a framework that treats story-to-audio generation as a precise orchestration and signal processing problem. This framework is enabled without ground-

Accurate breast cancer risk prediction from screening mammography is critical for enabling personalized screening intervals and early detection. Recent deep learning methods have shown the value of longitudinal data and explicit temporal alignment. However, existing approaches either perform explicit alignment using a single mammographic view or model multiple views without explicit longitudinal alignment, limiting their ability to exploit the complementary spatial-temporal information used in c

Ant Group’s AI Safety Lab has open-sourced SingGuard-NSFA, a safety guardrail model for autonomous agents, and disclosed details of SingGuard, a multimodal safety model. SingGuard-NSFA is designed to detect risks such as prompt injection, sensitive data theft, malicious code execution, resource abuse, and permission misuse before agents take action. The model covers seven major risk […]

Post-training is essential for refining the domain-specific capabilities of large language models (LLMs), yet existing reward optimization and distribution matching methods tightly couple policy exploration with distribution alignment. This coupling forces expensive exploration directly on the policy model and severely hinders the asynchronous generation, reuse, and cross-model transfer of optimization signals. In this paper, we propose Proxy-guided Update Signal Transfer (PUST), a novel post-tr

Learning, at its core, extends beyond memorization to the ability to reason and solve novel problems by navigating a space of possibilities. We introduce Lumo-2, a latent world-action model that generates actions by reasoning over world dynamics in latent space. The learned latent world dynamics capture physically grounded visual transitions, naturally encoding future possibilities and providing a unified substrate for cross-modal alignment. This formulation enables predictive reasoning akin to

Decision support systems (DSS) increasingly run retention what-if analysis on synthetic customer populations, because privacy constraints preclude unrestricted use of real data. Such a system is trustworthy only if the synthetic data lead managers to the same decisions as the real data would; yet prevailing criteria certify distributional similarity, not decision alignment, so a synthetic population can match every marginal distribution while still steering a marketing team toward the wrong camp

Safety evaluation of large language models (LLMs) relies largely on single-turn attack datasets and single-judge scoring, underestimating risk from adaptive multi-turn adversaries and reporting a single success rate that does not separate partially actionable outputs from those carrying complete operational detail. We propose AMT-X (Adaptive Multi-Turn Exploitation), a phase-structured multi-turn red-teaming framework. Unlike prior multi-turn attacks that rely on ad hoc escalation or free-form p

Modern large language models (LLMs) operate in interactive multi-turn settings, making multi-turn jailbreaking a realistic threat model and an important setting for automated red teaming. A core challenge in learning multi-turn jailbreak attackers is credit assignment: different turns contribute differently to the final outcome, yet existing learning signals are often too coarse to identify their individual contributions. We propose decomposed credit GRPO (DC-GRPO), a unified turn-level credit a

Face super-resolution is the task of increasing the resolution of an image containing a face thereby adding finer detail. It is a ubiquitous task in many computer vision applications and quite often the user isn't even aware that it is being performed. However, doing it with high fidelity is challenging as it is an ill-posed problem. In this paper we present a reference-based solution for face super-resolution that uses higher resolution reference images to aid in the task. We show an alignment

A former Mayo Clinic research director claims she was silenced, demoted and ultimately fired for sounding the alarm on AI safety and patient privacy lapses at the health system. Traci Tamiko Eto is now suing Mayo for retaliation. The post Former Mayo Clinic Leader Sues System Over Alleged AI Cover-Up: 6 Things to Know appeared first on MedCity News .

Decoding brain activity is useful for characterizing brain processes and understanding the functional architecture underlying cognition. However, the inter-individual variability in brain response patterns limits the development of decoders that generalize across individuals. A solution to this challenge is functional alignment: aligning functional data across individuals before training population-level decoders. The core issue is to strike the balance between aligning functional features and p