Search: “safety”

Eight key points from the most recent essay in the “AI as Normal Technology” series by @sayashk and me. Do AI Risks Require Extraordinary Go…

Eight key points from the most recent essay in the “AI as Normal Technology” series by @sayashk and me. Do AI Risks Require Extraordinary Government Intervention? 1. There is general consensus that AI

safetyyann-lecun--x

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

Governance by Design: Architecting Agentic AI for Organizational Learning and Scalable Autonomy

DGX agent

arXiv:2605.20210v1 Announce Type: cross Abstract: Agentic AI systems - systems that can pursue goals through multi-step planning and tool-mediated action with limited direct supervision - are moving f

safetyarxiv-cs-ai

22 May 2026

How can reasoning capability empower the AI copilot robot in endoscopic surgery

arXiv:2605.22322v1 Announce Type: new Abstract: Reasoning capability has significantly advanced complex logical inference and robotic decision-making in general domains. However, its potential in the

Making the Discrete Continuous: Synthetic RAW Augmentations for Fine-Grained Evaluation of Person Detection Performance in Low Light

arXiv:2605.22455v1 Announce Type: new Abstract: Real-world deployment of AI vision models is both fueled and limited by the data available for training and testing. Real datasets are sparse and uneven

Open-source LLMs administer maximum electric shocks in a Milgram-like obedience experiment

arXiv:2605.21401v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed as autonomous agents that make sequences of decisions over extended interactions in high-stakes

Parallel OctoMapping: A Scalable Framework for Enhanced Path Planning in Autonomous Navigation

arXiv:2603.22508v2 Announce Type: replace Abstract: Mapping is essential in robotics and autonomous systems because it provides the spatial foundation for path planning. Efficient mapping enables plan

Sources: last-minute calls with Elon Musk, Mark Zuckerberg, David Sacks, and others helped persuade President Trump not to sign the highly anticipated AI EO (Washington Post)

Washington Post: Sources: last-minute calls with Elon Musk, Mark Zuckerberg, David Sacks, and others helped persuade President Trump not to sign the highly anticipated AI EO — Industry leaders warned

safetytechmeme

update @thestalwart found more recent models less vulnerable. would be good to do a broad study of this.

Gary Marcus notes that researcher @thestalwart has found more recent AI models to be less vulnerable to certain attacks or exploits, and suggests that a comprehensive study across multiple models woul

safetygary-marcus--x

AI-Assisted Competency Assessment from Egocentric Video in Simulation-Based Nursing Education

arXiv:2605.20233v1 Announce Type: new Abstract: Assessing learner competency in clinical simulation requires expert observation that is time-intensive, difficult to scale, and subject to inter-rater v

AIMBio-Mat: An AI-Native FAIR Platform for Closed-Loop Materials Discovery and Biomedical Translation

arXiv:2605.21083v1 Announce Type: cross Abstract: Materials discovery and biomedical translation increasingly require models that can reason across composition, processing, structure, biological respo

CHEM: Estimating and Understanding Hallucinations in Deep Learning for Image Processing

arXiv:2512.09806v2 Announce Type: replace Abstract: Deep learning-based methods have recently achieved significant success in image reconstruction problems. However, challenges have emerged, as these

Do No Harm? Hallucination and Actor-Level Abuse in Web-Deployed Medical Large Language Models

arXiv:2605.20591v1 Announce Type: new Abstract: Medical large language models (LLMs), including custom medical GPTs (MedGPTs) and open-source models, are increasingly deployed on web platforms to prov

safetyarxiv-cs-cl

FineVision: Open Data Is All You Need

arXiv:2510.17269v2 Announce Type: replace Abstract: The advancement of vision-language models (VLMs) is hampered by a fragmented landscape of inconsistent and contaminated public datasets. We introduc

GAMR: Geometric-Aware Manifold Regularization with Virtual Outlier Synthesis for Learning with Noisy Labels

arXiv:2605.20727v1 Announce Type: new Abstract: Deep neural networks (DNNs) experience significant performance degradation when processing noisy labels, primarily due to overfitting on mislabeled data

GraphDiffMed: Knowledge-Constrained Differential Attention with Pharmacological Graph Priors for Medication Recommendation

arXiv:2605.20188v1 Announce Type: new Abstract: Recommending safe and effective medication combinations from electronic health records (EHRs) is a core clinical AI problem, yet it remains difficult be

Lost in Fog: Sensor Perturbations Expose Reasoning Fragility in Driving VLAs

arXiv:2605.21446v1 Announce Type: new Abstract: Interpretable autonomous driving planners depend not only on generating explanations, but also on those explanations remaining reliable under real-world

Mechanistic Interpretability for Learning Assurance of a Vision-Based Landing System

arXiv:2605.20607v1 Announce Type: cross Abstract: EASA's learning-assurance guidance requires data-driven aviation systems to build and monitor their own situation representation, yet for neural netwo

Multi-Agent Reinforcement Learning for Safe Autonomous Driving Under Pedestrian Behavioral Uncertainty

arXiv:2605.20255v1 Announce Type: new Abstract: Simulation-based testing of self-driving cars (SDCs) typically relies on scripted or simplified pedestrian models that do not capture the heterogeneity

Preference-aware Influence-function-based Data Selection Method for Efficient Fine-Tuning

arXiv:2605.21422v1 Announce Type: new Abstract: As LLMs continue to scale, improving training efficiency increasingly depends on using data more effectively. Data selection addresses this problem by a

Quadratic Characterizations for Reachability Analysis of Neural Networks

arXiv:2605.20482v1 Announce Type: new Abstract: Quadratic constraints (QCs) are widely used to characterize nonlinearities and uncertainties, but generic analytical characterizations can be conservati

Spacetime Optimal-Transport Attention for Visuo-Haptic Imitation Learning of Contact-Rich Manipulation

arXiv:2605.20433v1 Announce Type: new Abstract: Contact-rich manipulation tasks such as tight-clearance insertion, connector mating, polishing, and surface-conforming wiping remain difficult for data-

SurgOnAir: Hierarchy-Aware Real-Time Surgical Video Commentary

arXiv:2605.21132v1 Announce Type: new Abstract: Understanding surgical workflow in real time is fundamental for intelligent surgical embodiment, where AI systems continuously perceive and respond as s

Trusted Weights, Treacherous Optimizations? Optimization-Triggered Backdoor Attacks on LLMs

arXiv:2605.20641v1 Announce Type: cross Abstract: Inference optimization is a vital technique for deploying LLMs at scale. Compilation is the most widely adopted optimization technique for LLMs. While

When AI Gets it Wrong: Reliability and Risk in AI-Assisted Medication Decision Systems

arXiv:2604.01449v3 Announce Type: replace-cross Abstract: Artificial intelligence (AI) systems are increasingly integrated into healthcare and pharmacy workflows, supporting tasks such as medication r

ARC-RL: A Reinforcement Learning Playground Inspired by ARC Raiders

arXiv:2605.19503v1 Announce Type: cross Abstract: Reinforcement learning for legged locomotion has matured into a stack of multi-component reward functions and physics-engine benchmarks whose morpholo

Attention-Guided Reward for Reinforcement Learning-based Jailbreak against Large Reasoning Models

arXiv:2605.19485v1 Announce Type: new Abstract: Large Reasoning Models (LRMs) have demonstrated remarkable capabilities in solving complex problems by generating structured, step-by-step reasoning con

Backtracking When It Strays: Mitigating Dual Exposure Biases in LLM Reasoning Distillation

arXiv:2605.19433v1 Announce Type: cross Abstract: Large language models (LLMs) have achieved remarkable success in complex reasoning tasks via long chain-of-thought (CoT), yet their immense computatio

Be Kind, Rewrite: Benign Projections via Rewriting Defend Against LLM Data Poisoning Attacks

arXiv:2605.19147v1 Announce Type: cross Abstract: Large language models (LLMs) are highly susceptible to backdoor attacks (BAs), wherein training samples are poisoned using trigger-based harmful conte

Model Releases

Beyond Imitation: Learning Safe End-to-End Autonomous Driving from Hard Negatives

model-releasesarxiv-cs-cv

arXiv:2605.19771v1 Announce Type: cross Abstract: Existing imitation learning methods for end-to-end autonomous driving predominantly learn from successful demonstrations by minimizing geometric devia

⚠️👇 🚨Breaking ⚠️ If we can’t make AI agents follow rules, we are screwed. New study from METR reports that “when the agents were faced wit…

⚠️👇 🚨Breaking ⚠️ If we can’t make AI agents follow rules, we are screwed. New study from METR reports that “when the agents were faced with hard tasks, they routinely violated constraints” This—routin

safetygary-marcus--x

CADENet: Condition-Adaptive Asynchronous Dual-Stream Enhancement Network for Adverse Weather Perception in Autonomous Driving

arXiv:2605.19837v1 Announce Type: cross Abstract: Adverse weather (rain, fog, sand, and snow) degrades camera-based object detection in autonomous vehicles. Existing enhancement-then-detect approaches

CEPO: RLVR Self-Distillation using Contrastive Evidence Policy Optimization

arXiv:2605.19436v1 Announce Type: cross Abstract: When a model produces a correct solution under reinforcement learning with verifiable rewards (RLVR), every token receives the same reward signal rega

safetyarxiv-cs-cl

Compliant Explicit Reference Governor for Contact Friendly Robotic Manipulators

arXiv:2504.09188v2 Announce Type: replace Abstract: This paper introduces the Compliant Explicit Reference Governor (CERG), a modular reference management system that enables robots to interact physic

Model Releases

Contrastive Reasoning Alignment: Reinforcement Learning from Hidden Representations

model-releasesarxiv-cs-ai

arXiv:2603.17305v2 Announce Type: replace Abstract: We propose CRAFT, a red-teaming alignment framework that leverages model reasoning capabilities and hidden representations to improve robustness aga

ESLD (External Surrogate Latent Defense): A Latent-Space Architecture for Faster, Stronger Prompt-Injection Defense

arXiv:2605.18918v1 Announce Type: cross Abstract: Modern AI assistants are agentic. To answer a single user request, the underlying language model pulls in information from many sources, such as web s

Model Releases

Explainable Wastewater Digital Twins: Adaptive Context-Conditioned Structured Simulators with Self-Falsifying Decision Support

model-releasesarxiv-cs-ai

arXiv:2605.19826v1 Announce Type: new Abstract: Operators of safety-critical industrial processes increasingly rely on digital twins to screen control interventions, but such simulators rarely carry c

Graph Neural Planning and Predictive Control for Multi-Robot Communication-Constrained Unlabeled Motion Planning

arXiv:2605.19209v1 Announce Type: new Abstract: The multi-robot unlabeled motion planning problem of concurrently assigning robots to goals and generating safe trajectories is central in many collabor

Hamilton--Jacobi Reachability for Spacecraft Collision Avoidance

arXiv:2605.20138v1 Announce Type: new Abstract: This article presents a Hamilton--Jacobi (HJ) reachability framework for a two--satellite collision avoidance problem operating in the same circular orb

Local Ai

KIO-planner: Attention-Guided Single-Stage Motion Planning with Dual Mapping for UAV Navigation

arXiv:2605.19703v1 Announce Type: new Abstract: Autonomous UAV flight in confined, wall-dense environments requires low-latency and reliable motion planning under strict safety constraints. Traditiona

local-aiarxiv-cs-ro

Knowing When Not to Predict: Self Supervised Learning and Abstention for Safer DR Screening

arXiv:2605.19133v1 Announce Type: cross Abstract: Self-supervised learning (SSL) is now a standard way to pretrain medical image models, but performance is still mostly judged by downstream accuracy.

Position: Uncertainty Quantification in LLMs is Just Unsupervised Clustering

arXiv:2605.19220v1 Announce Type: cross Abstract: Uncertainty Quantification (UQ) is widely regarded as the primary safeguard for deploying Large Language Models (LLMs) in high-stakes domains. However

Probabilistic Recursively Feasible Motion Planning Under Uncertain Environments

arXiv:2605.19015v1 Announce Type: cross Abstract: Safe motion planning in uncertain, time-varying environments is challenging because the safe region can change unpredictably across planning steps, of

Pseudocode-Guided Structured Reasoning for Automating Reliable Inference in Vision-Language Models

arXiv:2605.19663v1 Announce Type: new Abstract: Vision-Language Models (VLMs) are becoming the cornerstone of high-level reasoning for robotic automation, enabling robots to parse natural language com

Robotics-Inspired Guardrails for Foundation Models in Socially Sensitive Domains

arXiv:2605.19940v1 Announce Type: new Abstract: Foundation models are increasingly deployed in socially sensitive domains such as education, mental health, and caregiving, where failures are often cum

Augmenting Human Evaluation with LLM Judges: How Many Human Reviews Do You Need?

arXiv:2605.16354v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used as automated evaluators of AI systems, including in high-stakes applications. In this role, LLMs ar

Code as Agent Harness

arXiv:2605.18747v1 Announce Type: cross Abstract: Recent large language models (LLMs) have demonstrated strong capabilities in understanding and generating code, from competitive programming to reposi

Ethical Hyper-Velocity (EHV): A Provably Deterministic Governance-Aware JIT Compiler Architecture for Agentic Systems

arXiv:2605.17909v1 Announce Type: new Abstract: As autonomous agentic systems scale across regulated critical infrastructures, the lack of mechanistic, hardware-rooted enforcement for high-frequency p

Event-Grounded Sparse Autoencoders for Vision-Language-Action Policies

arXiv:2605.17204v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) policies translate language and visual inputs into robot actions, where their hidden representations directly shape close