safety

Selective Token-Level Cryptographic Redaction for Privacy-Preserving Clinical Deployment of Large Language Models

arXiv:2606.03399v1 Announce Type: new Abstract: While large language models (LLMs) are increasingly used for clinical applications, many existing pipelines require sending raw sensitive health informa

Safety

Self-Refining Agentic Reinforcement Learning for Vision-Conditioned UAV Navigation

DGX agent

arXiv:2606.03963v1 Announce Type: cross Abstract: Deep reinforcement learning has shown strong potential for enabling autonomous robots to learn complex navigational tasks. However, its practical use

safetyarxiv-cs-ai

3 Jun 2026

SeSE: Black-Box Uncertainty Quantification for Large Language Models Based on Structural Information Theory

arXiv:2511.16275v4 Announce Type: replace-cross Abstract: Reliable uncertainty quantification (UQ) is essential for deploying large language models (LLMs) in safety-critical scenarios, as it enables t

SimuScene: Simulation-Ready Compositional 3D Scene Reconstruction from a Single Image

arXiv:2606.03994v1 Announce Type: new Abstract: Reconstructing interactive, simulation-ready 3D scenes from a single image is a critical bottleneck for robotic manipulation. While recent single-image

SkelHCC: A Hyperbolic CLIP-Driven Cache Adaptation Framework for Skeleton-based One-Shot Action Recognition

arXiv:2606.03610v1 Announce Type: new Abstract: Skeleton-based action recognition aims to understand human behaviors from body joint sequences and is especially challenging in the one-shot setting, wh

Sparse-View Lung Nodule Volumetry from Digitally Reconstructed Radiographs via AReT: Anatomy-Regularized TensoRF

arXiv:2606.02639v1 Announce Type: cross Abstract: We identify and resolve a previously unreported failure mode in TensoRF when applied to X-ray attenuation fields: the default density shift of -10, or

Spectral Asymptotics of Neural Network Loss Landscapes: An Exact Decomposition of the Curvature Exponent

arXiv:2606.02596v1 Announce Type: new Abstract: The curvature exponent alpha in h_k propto sigma_k^alpha -- governing how Hessian eigenvalues scale with gradient singular values -- varies systematical

SplitAdapter: Load-Aware Humanoid Loco-Manipulation via Factorized Adaptation

arXiv:2606.03297v1 Announce Type: new Abstract: Humanoid loco-manipulation requires stable whole-body control under varying object masses and pickup/placement heights. This becomes particularly challe

Strongly Polynomial Time Complexity of Policy Iteration for L_infty Robust MDPs

arXiv:2601.23229v2 Announce Type: replace Abstract: Markov decision processes (MDPs) are a fundamental model in sequential decision making. Robust MDPs (RMDPs) extend this framework by allowing uncert

Suno - a company that trained on 'essentially all music files of reasonable quality that are accessible on the open Internet', and argues it…

Suno - a company that trained on 'essentially all music files of reasonable quality that are accessible on the open Internet', and argues it does not need to pay to do so - is now valued at $5.4 billi

Supercell, King, and Sybo warn that EU's Digital Fairness Act, requiring pop-ups showing real-world values of virtual currencies, could make games 'unplayable' (Richard Milne/Financial Times)

Richard Milne / Financial Times: Supercell, King, and Sybo warn that EU's Digital Fairness Act, requiring pop-ups showing real-world values of virtual currencies, could make games “unplayable” — Maker

safetytechmeme

Taiji: Pareto Optimal Policy Optimization with Semantics-IDs Trade-off for Industrial LLM-Enhanced Recommendation

arXiv:2606.03866v1 Announce Type: cross Abstract: Scaling recommender systems via large language models (LLMs) has become a prominent trend in the industry. However, aligning the LLM's semantic space

Temporal Action Selection for Action Chunking

arXiv:2511.04421v2 Announce Type: replace Abstract: Action chunking is a widely adopted approach in Learning from Demonstration (LfD). By modeling multi-step action chunks rather than single-step acti

TGV-KV: Text-Grounded KV Eviction for Vision-Language Models

arXiv:2606.03075v1 Announce Type: new Abstract: Vision-Language Models (VLMs) inherit the auto-regressive generation paradigm and cache the keys and values (KV) of all previous tokens to accelerate in

The Ghost Annotator: a Framework to Explore Human Label Variation in Content Moderation through Conformal Prediction

arXiv:2606.02911v1 Announce Type: new Abstract: Current research primarily focuses on model performance, while comparatively less attention has been devoted to uncertainty estimation, particularly in

The Shadow Price of Reasoning: Economic Perspective on Optimal Budget Allocation for LLMs

arXiv:2606.03092v1 Announce Type: new Abstract: Inference-time scaling has emerged as a critical avenue for enhancing Large Language Models' performance, yet real-world deployment is constrained by st

Think-Before-Speak: From Internal Evaluation to Public Expression in Multi-Agent Social Simulation

arXiv:2606.03137v1 Announce Type: new Abstract: LLM-based multi-agent simulation offers a promising way to study social interaction, deliberation, and collective opinion dynamics. However, many existi

Too Much of a Good Thing: When sim2real Efforts Impede Policy Learning (And What to Do About It)

arXiv:2606.02636v1 Announce Type: cross Abstract: While sim2real efforts are necessary for effective policy transfer to hardware, there is such a thing as too much of a good thing. We argue that sim2r

Tool-Aware Optimization with Entropy Guidance for Efficient Agentic Reinforcement Learning

arXiv:2606.03762v1 Announce Type: cross Abstract: Agentic reinforcement learning (RL) equips large language models (LLMs) with tool-use capabilities that substantially improve reasoning on complex tas

Towards a Science of AI Agent Reliability

arXiv:2602.16666v3 Announce Type: replace Abstract: AI agents are increasingly deployed to execute important tasks. While rising accuracy scores on standard benchmarks suggest rapid progress, many age

Train Once, Reuse Everywhere: Generalizable Implicit In-Context Learning by Routing Attention

arXiv:2509.22854v2 Announce Type: replace Abstract: Implicit in-context learning (ICL) has newly emerged as a promising paradigm that simulates ICL behaviors in the representation space of large langu

Trust Grok

Trust Grok Yes. Racism toward white people exists—prejudice and discrimination based on race, full stop. The redefinition that limits it to 'power + prejudice' is ideological sleight-of-hand designed

safetyelon-musk--x

TTT-VLA: Test-Time Latent Prompt Optimization for Vision-Language-Action Models

arXiv:2606.03127v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models trained on large-scale data have made remarkable progress, but they remain vulnerable to distribution shifts at depl

Unified Video-Action Joint Denoising for Dexterous Action and Data Generation

arXiv:2606.03868v1 Announce Type: new Abstract: Recent world action models leverage video foundation models by aligning broad visual-dynamics priors with executable robot actions. We revisit this alig

UnsOcc: 3D Semantic Occupancy Prediction in Unstructured Scene via Rendering Fusion

arXiv:2606.03581v1 Announce Type: new Abstract: Unstructured scenes present unique challenges for autonomous driving, as irregular obstacles and sparse scene layouts undermine the effectiveness of tra

Using Reward Uncertainty to Induce Diverse Behaviour in Reinforcement Learning

arXiv:2606.03962v1 Announce Type: cross Abstract: Classical reinforcement learning (RL) typically seeks a deterministic policy that maximizes the expected sum of a scalar reward. Yet, modern applicati

Validation-Gated Multi-Agent Governance for Online Adaptation of Thermal-Hydraulic Surrogate Models under Operating-Regime Shift

arXiv:2606.03321v1 Announce Type: new Abstract: Artificial-intelligence surrogates can support second-by-second thermal-hydraulic forecasting, but models selected and frozen offline may become conditi

We're building a global movement to prohibit superintelligence internationally. Our new campaign in Canada is supported by over 30 MPs and S…

We're building a global movement to prohibit superintelligence internationally. Our new campaign in Canada is supported by over 30 MPs and Senators calling for a ban. After our success in the UK, it's

safetyconnor-leahy--x

What are you investing in? Hopium:

What are you investing in? Hopium: Former BlackRock fund manager Ed Dowd on the stock market: 'If 45% of your market cap is AI and there's no profits yet, what are you investing in?' 'You're investing

When Attention Collapses: Stage-Aware Visual Token Pruning from Structure to Semantics

arXiv:2606.03569v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) have demonstrated remarkable capabilities but suffer from significant computational overhead during inference. While vis

When Graph Tokens Sink: A Mechanistic Analysis of Graph Language Models

arXiv:2606.03712v1 Announce Type: new Abstract: Graph Language Models (GLMs) have become a promising direction for adapting Large Language Models (LLMs) to graph learning tasks. By transforming graph

When Models Refuse: Political Steerability and Feature Richness as Measures of Ideological Depth

arXiv:2508.21448v3 Announce Type: replace Abstract: Large language models (LLMs) sometimes refuse to follow benign instructions, such as declining to argue a political position or adopt a stated perso

when sam quotes the bible, you know things aren’t going well for OpenAI

when sam quotes the bible, you know things aren’t going well for OpenAI one of the quotes i find most inspiring on a hard day: 'Whatever your hand finds to do, do it with all your might, for in the re

When to Re-Plan: Subgoal Persistence in Hierarchical Latent Reasoning

arXiv:2606.03741v1 Announce Type: new Abstract: Long-horizon reasoning requires a system to commit to medium-horizon intent without becoming rigid: re-plan too often and computation never coheres into

Who Deserves the Reward? SHARP: Shapley Credit-based Optimization for Multi-Agent System

arXiv:2602.08335v2 Announce Type: replace Abstract: Integrating Large Language Models (LLMs) with external tools via multi-agent systems offers a promising new paradigm for decomposing and solving com

World Models Meet Language Models: On the Complementarity of Concrete and Abstract Reasoning

arXiv:2606.03603v1 Announce Type: cross Abstract: World models and multimodal large language models (MLLMs) provide complementary capabilities for predicting future outcomes from static visual observa

yes

yes @GaryMarcus Like, is this based on ppl actually buying into Musk's truly unhinged pie-in-the-sky timelines for his fantastical sci-fi aspirations, which seem unlikely to even be technologically fe

You don’t need to be @garymarcus to know which way the wind blows.

You don’t need to be @garymarcus to know which way the wind blows. I would be way more bullish on AI if it actually worked and was actually replacing real humans at scale. Nothing is changing and we’r

A Modelling and Evaluation Framework for EuroCrops-Driven Sentinel-2 Crop Segmentation

arXiv:2606.00676v1 Announce Type: new Abstract: This work presents a configurable pipeline for generating semantic-segmentation-ready agricultural datasets from Sentinel-2 imagery and EuroCrops parcel

A Monosemantic Attribution Framework for Stable Interpretability in Clinical Neuroscience Transformer-Based Language Models

arXiv:2601.17952v2 Announce Type: replace-cross Abstract: Interpretability remains a key challenge for deploying language models (LM) in clinical settings such as progression diagnosis of Alzheimer di

A Practical Upper Bound on Selection Bias Effects in Medical Prediction Models

arXiv:2606.00563v1 Announce Type: cross Abstract: Selection bias is a common and often unavoidable aspect of real-world data that challenges the generalizability of machine learning models. When model

A Predictive Control Strategy to Offset-Point Tracking for Agricultural Mobile Robots

arXiv:2603.28439v2 Announce Type: replace Abstract: Robots are increasingly being deployed in agriculture to support sustainable practices and improve productivity. They offer strong potential to enab

A Protocol-Language Model for Network Intrusion (Without Deep Packet Inspection)

arXiv:2606.00155v1 Announce Type: cross Abstract: Modern network intrusion detection systems (NIDS) are caught in a structural contradiction: the protocols carrying the highest threat intelligence are

A Shared Valence Axis Across Modern LLMs and Human EEG: The Saturation Regularity

arXiv:2606.00129v1 Announce Type: cross Abstract: Large language models (LLMs) have emerged as powerful representation learners whose internal features increasingly align with human cognition. We stud

Adaptive Order Policies for Masked Diffusion

arXiv:2606.00295v1 Announce Type: new Abstract: Masked diffusion models have seen great success in capturing data distributions over discrete sequences in domains such as text and proteins. These mode

Adaptive Time Series Reasoning via Segment Selection

arXiv:2602.18645v2 Announce Type: replace Abstract: Time series reasoning tasks often start with a natural language question and require targeted analysis of a time series. Evidence may span the full

Addressing Longstanding Challenges in Cognitive Science with Language Models

arXiv:2511.00206v3 Announce Type: replace Abstract: Cognitive science faces ongoing challenges in research integration, formalization, conceptual clarity, and other areas, in part due to its multiface

Advancing youth safety and opportunity through global leadership

OpenAI outlines initiatives and commitments to enhance youth safety while creating educational and economic opportunities for young people globally. The effort emphasizes OpenAI's role in responsible

safetyopenai