safety

Diffusion Forcing Planner: History-Annealed Planning with Time-Dependent Guidance for Autonomous Driving

arXiv:2606.11019v1 Announce Type: cross Abstract: Learning-based motion planners, despite recent progress, often suffer from temporal inconsistency. Small perturbations across frames can accumulate in

discussed more here:

DGX agent

discussed more here: I was first to call the death of tokenmaxxing (ahead of @zerohedge), among the first (in 2023) to warn that the economics of Generative AI was problematic, and first (also 2023) t

safetygary-marcus--x

10 Jun 2026

Dissect and Prune: Enhancing Robustness in AI-Generated Image Detection

arXiv:2606.10309v1 Announce Type: new Abstract: While existing AI-generated image detectors report high performance, we identify that this is largely driven by a critical prediction asymmetry: a bias

Does Reasoning Preserve Alignment? On the Trustworthiness of Large Reasoning Models

arXiv:2606.11046v1 Announce Type: new Abstract: Instruction-tuned LLMs are increasingly converted into reasoning models through post-training to improve multi-step task performance. This conversion is

Dropout-GRPO: Variational Stochasticity for Continuous Latent Reasoning

arXiv:2606.10184v1 Announce Type: cross Abstract: Group Relative Policy Optimization (GRPO) relies on the diversity of K rollouts within each group; otherwise, the group-mean advantage A^{(k)} = r^{(k

Early-Token Confidence Predicts Reasoning Quality in Multi-Agent LLM Debate

arXiv:2606.10307v1 Announce Type: new Abstract: Evaluating reasoning quality in multi-agent LLM systems is challenging, especially for open-ended tasks without reference answers. We investigate whethe

EM-Fall: Embodied mmWave Sensing for Day-and-Night Fall Detection on Humanoid Robots

arXiv:2606.11109v1 Announce Type: new Abstract: Falls are one of the leading causes of injury and hospitalization among elderly individuals, making reliable fall awareness an essential capability for

Embodiment-conditioned Generalist Control for Multirotor Aerial Robots

arXiv:2606.10857v1 Announce Type: cross Abstract: We present a generalist position control policy capable of controlling arbitrary multirotor configurations of a certain rotor count (e.g., hexarotors

Encoding the Euler Characteristic Transform

arXiv:2606.10824v1 Announce Type: new Abstract: The Euler Characteristic Curve (ECC) records the Euler characteristic of a linearly embedded cell complex as a function of filtration height in a given

Enhancing Multilingual LLM-based ASR with Mixture of Experts and Dynamic Downsampling

arXiv:2606.10439v1 Announce Type: cross Abstract: The rapid progress of large language models (LLMs) has opened up a new frontier for automatic speech recognition (ASR), making their effective integra

ERAlign: Energy-based Representation Alignment of GNNs and LLMs on Text-attributed Graphs

arXiv:2606.10461v1 Announce Type: cross Abstract: Text-attributed Graphs (TAGs) incorporate textual node attributes with graph structures to describe rich relational semantics. Recent efforts to integ

Ethical and Technical Limits of Deepfake Speech Datasets

arXiv:2606.10911v1 Announce Type: cross Abstract: Claims about the robustness and fairness of deepfake speech detectors are only as credible as the datasets used to train and evaluate those systems. W

Event-Driven Reinforcement Learning Enables Long-Horizon Control in Semiconductor Fabrication

arXiv:2606.10705v1 Announce Type: cross Abstract: Reinforcement learning promises to optimize sequential decisions in large-scale systems. Semiconductor manufacturing systems are stochastic and highly

Every day I wake up hoping there is nothing new in AI that I feel like I need to write about. Just for one day. Every day I am disappointed.

Gary Marcus expresses frustration about the rapid pace of AI developments, noting that he hopes for a day without significant breakthroughs requiring commentary, but finds himself continually disappoi

Exactly

Exactly It’s not social media that’s “inflaming tensions”. It’s not Elon Musk. It’s not Nigel Farage. It’s not the ‘far-right’. It is the very deliberate policy of mass uncontrolled immigration & open

safetyelon-musk--x

experts on german law: can Google appeal, how soon does the new ruling take effect, etc?

This post likely discusses expert analysis on German legal proceedings involving Google, covering questions about appeal options available to the company, the timeline for implementing a court ruling,

Exploring Accurate and Transparent Domain Adaptation in Predictive Healthcare via Concept-Grounded Orthogonal Inference

arXiv:2602.12542v2 Announce Type: replace-cross Abstract: Deep learning models for clinical event prediction on electronic health records (EHR) often suffer performance degradation when deployed under

Fast and Highly Expressive Policy Learning for Offline Reinforcement Learning via Bootstrapped Flow Q-Learning

arXiv:2606.10613v1 Announce Type: cross Abstract: Diffusion-based Q-learning has emerged as a powerful paradigm for offline reinforcement learning, but its reliance on multi-step denoising makes both

Flow Control: Steering Vision-Language-Action Models with Simple Real-Time Inputs

arXiv:2606.10180v1 Announce Type: cross Abstract: We introduce flow control of vision-language-action (VLA) models, a simple and effective way to steer VLA actions in real-time through generic inputs,

Flow-DPPO: Divergence Proximal Policy Optimization for Flow Matching Models

arXiv:2606.11025v1 Announce Type: new Abstract: Recent work has demonstrated that online reinforcement learning (RL) can substantially improve the quality and alignment of flow matching models for ima

For Robotaxis, Safety Must Be Built In, Not Bolted On

A car pulls up to the curb. The app says, “Your ride is here.” No one’s in the driver’s seat. For people who live in one of the dozens of cities now hosting robotaxi services, this is already a realit

safetynvidia-blog

Generalized-CVO: Fast and Correspondence-Free Local Point Cloud Registration with Second Order Riemannian Optimization

arXiv:2606.10019v1 Announce Type: cross Abstract: We propose a fast and correspondence-free local point cloud registration method that leverages geometric surface structure and reproducing kernel Hilb

GHOST: Hierarchical Sub-Goal Policies for Generalizing Robot Manipulation

arXiv:2606.10025v1 Announce Type: cross Abstract: We present GHOST, a framework for learning visuomotor manipulation policies that generalize beyond the training distribution. GHOST factorizes control

Globally Localizing Lunar Rover in Pixels via Graph Alignment

arXiv:2606.10602v1 Announce Type: new Abstract: Precise rover localization is a prerequisite for autonomous lunar exploration, yet the absence of Global Navigation Satellite System (GNSS) signals and

Going with the Flow: Koopman Behavioral Models as Pseudo Planners for Visuo-Motor Dexterity

arXiv:2602.07413v3 Announce Type: replace Abstract: Contemporary visuo-motor dexterity models often rely on expressive policy classes with diffusion and transformer backbones to achieve strong perform

Gradient-Guided Reward Optimization for Inference-time Alignment

arXiv:2606.09635v1 Announce Type: cross Abstract: Ensuring the reliability of Large Language Models (LLMs) under distribution drift requires inference-time adaptation. While inference-time alignment m

GUI-AC: Enhancing Continual Learning in GUI Agents

arXiv:2606.10522v1 Announce Type: new Abstract: Graphical User Interfaces (GUIs) serve as the dominant medium for human-computer interaction, yet building GUI agents that generalize across the vast di

GUIDE: Goal-Initialized Directional Understanding for End-to-End Visual Navigation

arXiv:2606.10832v1 Announce Type: new Abstract: Learning-based visual navigation for legged robots typically relies on continuous goal updates from hierarchical state estimation to provide a persisten

GuideWalk: Learning Unified Autonomous Navigation and Locomotion for Humanoid Robots across Versatile Terrains

arXiv:2606.10449v1 Announce Type: new Abstract: Humanoid robots have achieved strong locomotion capabilities, but reliable navigation on versatile terrains remains challenging because obstacle avoidan

Hand-centric Human-to-Robot Trajectory Transfer from Video Demonstrations via Open-World Contact Localization

arXiv:2606.10743v1 Announce Type: new Abstract: Learning from human video demonstrations remains challenging due to noisy hand-object interactions, unseen objects with partial observation, and cross-e

Hidden Consensus:Preference-Validity Compression in Human Feedback

arXiv:2606.10569v1 Announce Type: cross Abstract: Standard RLHF pipelines often reduce heterogeneous human judgments into a single scalar reward target. We argue that this reduction can mis-measure al

Hierarchical Policies from Verbal and Egocentric Human Signals for Natural Human-Robot Interaction

arXiv:2606.10276v1 Announce Type: cross Abstract: For natural human-robot interaction, a robot must understand human intent expressed not only through language but also through nonverbal signals such

HiGR: Industrial-Scale Hierarchical Generative Slate Recommendation Framework in Tencent

arXiv:2512.24787v3 Announce Type: replace-cross Abstract: Slate recommendation, which presents users with a ranked item list in a single display, is ubiquitous across mainstream online platforms. Whil

Human-AI Coordination Zones: A Framework for Designing Human-in-the-Loop Experiences with Agentic AI

arXiv:2606.09848v1 Announce Type: cross Abstract: As generative and agentic AI becomes embedded in everyday products, practitioners face a persistent challenge: how to design human-AI coordination --

I was first to call the death of tokenmaxxing (ahead of @zerohedge), among the first (in 2023) to warn that the economics of Generative AI w…

I was first to call the death of tokenmaxxing (ahead of @zerohedge), among the first (in 2023) to warn that the economics of Generative AI was problematic, and first (also 2023) to warn that OpenAI co

IDEAL: In-DEpth ALignment Makes A Discrete Representation AutoEncoder

arXiv:2606.11096v1 Announce Type: new Abstract: Built on pretrained vision foundation models (VFMs), representation autoencoders (RAEs) have recently emerged as a promising approach for constructing s

🔔 IF Anthropic is for real about safety, the should pause, for one month, and show leadership. If they won’t pause, even briefly, we have a…

🔔 IF Anthropic is for real about safety, the should pause, for one month, and show leadership. If they won’t pause, even briefly, we have a different kind of answer about what is real and what is mark

If the new German court decision — to hold LLM companies liable for the false claims their LLMs made — become universal, effectively sidelin…

If the new German court decision — to hold LLM companies liable for the false claims their LLMs made — become universal, effectively sidelining generative AI until better techniques came along, would

IMPACT: Learning Internal-Model Predictive Control for Forceful Robotic Manipulation

arXiv:2606.10818v1 Announce Type: cross Abstract: Real-world robotic manipulation tasks often involve forceful interactions with the environment, such as using tools of varying weights, transporting o

Improving Adversarial Transferability on Vision-Language Pre-training Models via Surrogate-Specific Bias Correction

arXiv:2606.10571v1 Announce Type: cross Abstract: Adversarial examples reveal vulnerabilities in Vision-Language Pre-training (VLP) models and provide insights for improving robustness. A key property

Improving Text-Instance Alignment Of Foreground Conditioned Out-Painting Via Customized Concept Embedding

arXiv:2606.10892v1 Announce Type: cross Abstract: To showcase products, merchants often incur substantial costs creating high-quality display images. Foreground Conditioned Outpainting (FCO) meets thi

In addition to transparency, I now believe frontier models should face mandatory third-party testing for cyber, bio, and autonomy risks—with…

In addition to transparency, I now believe frontier models should face mandatory third-party testing for cyber, bio, and autonomy risks—with the power to block or revoke deployment of models that pose

safetydario-amodei--x

Informed Asymmetric Actor-Critic: Leveraging Privileged Signals Beyond Full-State Access

arXiv:2509.26000v3 Announce Type: replace Abstract: Asymmetric reinforcement learning leverages privileged information available during training to improve learning under partial observability. Existi

Inverse Probability Weighting and Age-of-Information Aggregation for Decentralized Federated Learning under Partial Reception

arXiv:2606.10774v1 Announce Type: new Abstract: Decentralized Federated Learning (DFL) over lossy wireless networks faces two key challenges: selection bias, where updates from poor-quality links are

Is Fairness Truly Fair? Towards Reliable Lipschitz Fairness in Multi-Task Learning via Fixed-exorpdfstring{elta}{delta} Alignment

arXiv:2606.10632v1 Announce Type: cross Abstract: Lipschitz-style individual fairness formalizes the idea that semantically similar examples should receive similar predictions, but its evaluation in m

It is not literally impossible to pause or slow down. It is a choice. But if the “frontier labs” continue to refuse to do, I strongly believ…

It is not literally impossible to pause or slow down. It is a choice. But if the “frontier labs” continue to refuse to do, I strongly believe it will eventually lead to tremendous harm. (Essentially e

It Takes One to Bias Them All: Breaking Bad with One-Shot GRPO

arXiv:2606.10931v1 Announce Type: new Abstract: Warning: This paper contains several toxic and offensive statements. Modern large language models (LLMs) are typically aligned through large-scale post-

LAFP: Preserving Latent Action Structure in Latent Policy Learning via Flow Matching

arXiv:2606.10517v1 Announce Type: new Abstract: Learning high-quality latent actions from large-scale unlabeled videos, coupled with limited real-world interaction data for training an action decoder,