safety

KC-3DGS: Kurtosis-Constrained Gaussian Splatting for High-Fidelity View Synthesis

arXiv:2606.03120v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) enables real-time novel view synthesis by representing scenes as collections of anisotropic Gaussians optimized via differe

Language Models Need Sleep: Learning to Self-Modify and Consolidate Memories

DGX agent

arXiv:2606.03979v1 Announce Type: cross Abstract: The past few decades have witnessed significant advances in the design of machine learning algorithms, from early studies on task-specific shallow mod

safetyarxiv-cs-ai

3 Jun 2026

LAP: An Agent-to-Instrument Protocol for Autonomous Science

arXiv:2606.03755v1 Announce Type: new Abstract: Autonomous science is moving from demonstration to infrastructure. Large language model agents now plan experiments, and self-driving laboratories execu

Large Language Models Are Overconfident in Their Own Responses

arXiv:2606.03437v1 Announce Type: new Abstract: Prior work has shown that instruction-tuned large language models (LLMs) are less well calibrated than their base pre-trained counterparts. However, lit

Latent Activation Editing: Inference-Time Refinement of Learned Policies for Safer Multirobot Navigation

arXiv:2509.20623v2 Announce Type: replace Abstract: Reinforcement learning has enabled significant progress in complex domains such as coordinating and navigating multiple quadrotors. However, even we

LC-SAC: Lyapunov-Constrained Soft Actor-Critic via Koopman Operator Theory for Trajectory Tracking and Stabilization

arXiv:2602.04132v4 Announce Type: replace-cross Abstract: Reinforcement Learning (RL) has achieved remarkable success in solving complex sequential decision-making problems. However, its application t

Learning Power Flow with Confidence: A Probabilistic Guarantee Framework for Voltage Risk

arXiv:2308.07867v4 Announce Type: replace-cross Abstract: The absence of formal performance guarantees in machine learning (ML) has limited its adoption for safety-critical power system applications,

Learning Self-Interpretation from Interpretability Artifacts: Training Lightweight Adapters on Vector-Label Pairs

arXiv:2602.10352v2 Announce Type: replace-cross Abstract: Self-interpretation methods prompt language models to describe their own internal states, but remain unreliable due to hyperparameter sensitiv

Learning to Bet for Horizon-Aware Anytime-Valid Testing

arXiv:2603.19551v2 Announce Type: replace-cross Abstract: We develop horizon-aware anytime-valid tests and confidence sequences for bounded means under a strict deadline N. Using the betting/e-process

Learning Unmasking Policies for Diffusion Language Models

arXiv:2512.09106v4 Announce Type: replace Abstract: Diffusion (Large) Language Models (dLLMs) now match the downstream performance of their autoregressive counterparts on many tasks, while holding the

Let There Be Light: Reflection, Refraction and Scattering for Neural Operators

arXiv:2606.03262v1 Announce Type: new Abstract: Neural operators learn mappings between infinite-dimensional function spaces and provide a data-driven surrogate modeling paradigm for parametric partia

Letting Tutor Personas Speak Up for LLMs: Learning Steering Vectors from Dialogue via Preference Optimization

arXiv:2602.07639v2 Announce Type: replace Abstract: With the emergence of large language models (LLMs) as a powerful class of generative artificial intelligence (AI), their use in tutoring has become

Leveraging BART to Assess CS1 C++ Programming Assignments using Rubric-based Criteria

arXiv:2606.03814v1 Announce Type: new Abstract: This paper investigates rubric-aware, multitask fine-tuning of transformer models for automated grading of introductory C++ programming assignments, wit

Libra: Efficient Resource Management for Agentic RL Post-Training

arXiv:2606.03077v1 Announce Type: cross Abstract: Reinforcement learning (RL) has become a standard post-training paradigm for large language models (LLMs), extending beyond preference alignment to co

Localized, High-resolution Geographic Representations with Slepian Functions

arXiv:2602.00392v2 Announce Type: replace Abstract: Geographic data is fundamentally local. Disease outbreaks cluster in population centers, ecological patterns emerge along coastlines, and economic a

LoCAtion: Long-time Collaborative Attention Framework for High Dynamic Range Video Reconstruction

arXiv:2603.14377v2 Announce Type: replace Abstract: Prevailing High Dynamic Range (HDR) video reconstruction methods are fundamentally trapped in a fragile alignment-and-fusion paradigm. While explici

Margin Play: A Multi-Agent System For Public Policy Analysis In The Brazilian Equatorial Margin

arXiv:2606.02614v1 Announce Type: cross Abstract: The Brazilian Equatorial Margin (BEM) is Brazil's next offshore oil frontier, with operations expected to begin in 2026 in the Foz do Amazonas basin.

Measuring Weak-to-Strong Legibility of Reasoning Models

arXiv:2603.20508v2 Announce Type: replace-cross Abstract: Reasoning language models (RLMs) and the intermediate chains of thought they emit play an increasingly central role in multi-agent setups such

MetaWorld: Scaling Multi-Agent Video World Model from Single-view Video Data

arXiv:2606.02753v1 Announce Type: cross Abstract: Video world models are a foundational generative technology for embodied AI and the Metaverse, yet existing approaches are inherently limited to a sin

MIND: Multi-rationale INtegrated Discriminative Reasoning Framework for Multi-modal Large Models

arXiv:2512.05530v2 Announce Type: replace Abstract: Recently, multimodal large language models (MLLMs) have been widely applied to reasoning tasks. However, they suffer from limited multi-rationale se

Mitigating False Credit Propagation: Probabilistic Graphical Reward Aggregation for Rubric-Based Reinforcement Learning

arXiv:2606.03361v1 Announce Type: new Abstract: Rubric-based rewards are increasingly used for open-ended language model post-training, but criterion-level scores are often aggregated as independent u

Multi-component Causal Tracing in Large Language Models

arXiv:2606.03085v1 Announce Type: cross Abstract: Causal tracing systematically intervenes on a large language model's (LLM's) internal representations to uncover and quantify the causal pathways link

Multi-Segment Attention: Enabling Efficient KV-Cache Management for Faster Large Language Model Serving

arXiv:2606.02964v1 Announce Type: cross Abstract: Large Language Model (LLM) inference relies on key-value (KV) caches to avoid redundant attention computation. While approximate KV cache retention te

Neural Networks Provably Learn Spectral Representations for Group Composition

arXiv:2606.02993v1 Announce Type: new Abstract: Understanding how structured internal structure emerges during neural network training is central to the study of deep learning. We investigate this phe

No one. No one in their right mind.

No one. No one in their right mind. SpaceX is losing money hand over fist, nearly 5bn last year. @SpaceX’s only profitable business is Starlink, but its new satellites can only be launched by Starship

safetygary-marcus--x

NVIDIA OmniDreams: Real-Time Generative World Model for Closed-Loop Autonomous Vehicle Simulation

arXiv:2606.03159v1 Announce Type: cross Abstract: As autonomous vehicle capabilities advance, the safe evaluation of driving policies in long-tail scenarios remains a critical bottleneck. In closed-lo

OMP: One-step Meanflow Policy with Directional Alignment

arXiv:2512.19347v3 Announce Type: replace Abstract: Robot manipulation has increasingly adopted data-driven generative policy frameworks, yet the field faces a persistent trade-off: diffusion models s

OpenAI diverges from Trump's AI EO in a new policy paper, proposing cyber risk evaluations for advanced AI systems be mandatory and led by CAISI, not the NSA (Brendan Bordelon/Politico)

Brendan Bordelon / Politico: OpenAI diverges from Trump's AI EO in a new policy paper, proposing cyber risk evaluations for advanced AI systems be mandatory and led by CAISI, not the NSA — OpenAI's ne

safetytechmeme

OpenAI public policy agenda

OpenAI's public policy agenda outlines the organization's priorities and positions regarding AI regulation and governance. The document likely details OpenAI's recommendations for government policies,

safetyopenai

PAND: Prompt-Aware Neighborhood Distillation for Lightweight Fine-Grained Visual Classification

arXiv:2602.07768v3 Announce Type: replace-cross Abstract: Distilling knowledge from large Vision-Language Models (VLMs) into lightweight networks is crucial yet challenging in Fine-Grained Visual Clas

Pathway-Structured Privileged Distillation for Deployable Computational Pathology

arXiv:2606.02877v1 Announce Type: new Abstract: Integrating transcriptomics and histopathology can improve cancer risk modelling, yet practical use is constrained by the limited availability of RNA pr

Pextsuperscript{2}-DPO: Grounding Hallucination in Perceptual Processing via Calibration Direct Preference Optimization

arXiv:2606.03376v1 Announce Type: cross Abstract: Hallucination has recently garnered significant research attention in Large Vision-Language Models (LVLMs). Direct Preference Optimization (DPO) aims

PHAF-Personalized Hand Avatars in a Flash

arXiv:2606.03420v1 Announce Type: new Abstract: We present PHAF-Personalized Hand Avatars in a Flash, a personalized photo-realistic hand avatar which provides high quality multi-view renders from jus

PHASE: Physiology-Aware Hyperspectral Reconstruction via Object-to-Human Domain Adaptation

arXiv:2511.13020v2 Announce Type: replace-cross Abstract: Although hyperspectral imaging offers unparalleled non-invasive physiological insight, its bulky hardware, slow acquisition, and regulatory bu

Physical Plausibility Reasoning via HCM-GRPO: Empowering Compact Model for Superior Performance

arXiv:2511.10055v2 Announce Type: replace Abstract: The performance of image generation has been significantly improved in recent years. However, the study of image screening is rare, and its performa

Physics-Guided Policy Optimization with Self-Distillation

arXiv:2606.03620v1 Announce Type: cross Abstract: Self-distilled policy optimization (SDPO) has become a popular paradigm for LLM post-training, where a model learns from its own predictions condition

Planning with Uncertainty: Symmetries, Policy Inference, and Solution Compression

arXiv:2403.19883v2 Announce Type: replace Abstract: Fully-observable non-deterministic (FOND) planning is at the core of artificial intelligence planning with uncertainty. It models uncertainty throug

Post-Hoc Robustness for Model-Based Reinforcement Learning

arXiv:2606.03521v1 Announce Type: cross Abstract: To improve the real-world applicability of reinforcement learning (RL), the field of adversarially robust RL studies how to train agents under adversa

Preference-Calibrated Human-in-the-Loop Reinforcement Learning for Robotic Manipulation

arXiv:2606.03949v1 Announce Type: new Abstract: Human-in-the-loop reinforcement learning (HIL-RL) improves sample efficiency in real-robot manipulation through online human intervention. However, succ

PrimeSVT: An Automated Memory-aware Pruning Framework with Prioritized Compression Policy for Spiking Vision Transformers

arXiv:2606.03428v1 Announce Type: cross Abstract: The large sizes of Spiking Vision Transformers (SViTs) still hinder their embedded implementation, highlighting the need for model compression. State-

PsychoPass: Geometric Profiling of Multi-Turn Adversarial LLM Conversations

arXiv:2606.03136v1 Announce Type: cross Abstract: Multi-turn jailbreak attacks on large language models (LLMs) reveal a mismatch in current guardrails: they operate on individual turns, while attacks

Quantifying Faithful Confidence Expression in Large Reasoning Models

arXiv:2606.03969v1 Announce Type: cross Abstract: Reliable uncertainty communication is critical to the trustworthiness of LLMs, yet faithful calibration (FC)--the alignment between models' intrinsic

QUBRIC: Co-Designing Queries and Rubrics for RL Beyond Verifiable Rewards

arXiv:2606.03968v1 Announce Type: cross Abstract: Rubric-based RL is a promising route for extending reinforcement learning beyond verifiable rewards, yet existing methods optimize rubrics while treat

Rethinking Neural Width for Alternating Current Optimal Power Flow Proxies

arXiv:2606.03125v1 Announce Type: new Abstract: Deep learning proxies for Alternating Current Optimal Power Flow (ACOPF) lack systematic methods for determining architectural size. This paper conducts

Right Makes Might: Aligning Verified Hidden States Empowers RL Reasoning

arXiv:2606.03234v1 Announce Type: new Abstract: Reinforcement Learning from Verifiable Rewards (RLVR) has become the dominant approach for improving mathematical reasoning in large language models, ye

RoboCade: Gamifying Robot Data Collection

arXiv:2512.21235v3 Announce Type: replace Abstract: Imitation learning from human demonstrations has become a dominant approach for training autonomous robot policies. However, collecting demonstratio

Sam Altman swearing to tell the whole truth to the US Senate an hour or so before he lied his fanny off about caring about artists and creat…

Sam Altman swearing to tell the whole truth to the US Senate an hour or so before he lied his fanny off about caring about artists and creators and wanting them to get a fair shake. Hey OpenAI, when a

safetygary-marcus--x

See Less, Specify More: Visual Evidence Budgets for Generalizable VLAs

arXiv:2606.02735v1 Announce Type: cross Abstract: Generalization remains a central bottleneck for vision-language-action (VLA) models: under distractors, appearance shifts, and semantically similar ta