safety

On Optimizing Multimodal Jailbreaks for Spoken Language Models

arXiv:2603.19127v2 Announce Type: replace Abstract: As Spoken Language Models (SLMs) integrate speech and text modalities, they inherit the safety vulnerabilities of their LLM backbone while introduci

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Online Generation of Collision-Free Trajectories in Dynamic Environments

DGX agent

arXiv:2603.00759v2 Announce Type: replace Abstract: In this paper, we present an online method for converting an arbitrary geometric path, represented by a sequence of states, and generated by any pla

safetyarxiv-cs-ro

1 Jul 2026

PA-VAD: Diffusion-Based Pseudo-Only Video Anomaly Detection via Domain-Aligned Memory Updates

arXiv:2512.06845v2 Announce Type: replace Abstract: Deploying video anomaly detection (VAD) in the real world is often constrained by the scarcity, privacy, and cost of collecting real abnormal footag

Paper2Rebuttal: A Multi-Agent Framework for Transparent Author Response Assistance

arXiv:2601.14171v2 Announce Type: replace Abstract: Writing effective rebuttals is a high-stakes task that demands more than linguistic fluency, as it requires precise alignment between reviewer inten

Personalizing Marketplace Policies with Competing Objectives and Constrained Experiments: Evidence from a Job Marketplace

arXiv:2606.30932v1 Announce Type: new Abstract: Two-sided marketplaces connect distinct user groups whose interests often conflict -- improving outcomes on one side could degrade the other side's expe

PiLoT v2: Pixel-to-Orthogonal Map Alignment for Free-view UAV Geo-localization

arXiv:2606.31098v1 Announce Type: new Abstract: Real-time, drift-free UAV geo-localization is essential for autonomous missions in GNSS-denied environments. The pioneering system, PiLoT, achieves high

Policy Optimization Achieves Data-Dependent Regret Bounds in MDPs with Unknown Transitions

arXiv:2606.31769v1 Announce Type: new Abstract: We study policy optimization for online episodic tabular Markov decision processes with unknown transition kernels, aiming for best-of-both-worlds guara

PolicyGuard: From Organizational Policies to Neuro-SymbolicCompliance Review Engines

arXiv:2606.32004v1 Announce Type: new Abstract: Policy-grounded document review requires determining whether a target document complies with organization-specific policies, guidelines, or playbooks. W

Prompting Robot Teams with Natural Language

arXiv:2509.24575v2 Announce Type: replace-cross Abstract: This paper presents a framework to prompt multi-robot teams with high-level tasks using natural language expressions. Our objective is to use

QVal: Cheaply Evaluating Dense Supervision Signals for Long-Horizon LLM Agents

arXiv:2606.32034v1 Announce Type: cross Abstract: LLM agents increasingly act over long horizons, where a single trajectory can contain hundreds or thousands of actions. In these settings, outcome-onl

ReGRPO: Reflection-Augmented Policy Optimization for Tool-Using Agents

arXiv:2606.31392v1 Announce Type: new Abstract: Tool-augmented vision-language models (VLMs) can solve multimodal, multi-step tasks by calling external tools, yet they remain fragile in practice. Exis

Reinforcement Learning-Based Control for an Inline Skating Humanoid Robot

arXiv:2606.31807v1 Announce Type: new Abstract: As humanoid robots become increasingly dynamic, coupling them with reinforcement learning offers a promising approach to solving the complex, underactua

Reinforcement Learning with Metacognitive Feedback Elicits Faithful Uncertainty Expression in LLMs

arXiv:2606.32032v1 Announce Type: cross Abstract: Metacognition is a critical component of intelligence that describes the ability to monitor and regulate one's own cognitive processes. Yet LLMs exhib

Relational and Sequential Conformal Inference for Energy Time Series over Graphs via Foundation Models

arXiv:2606.31804v1 Announce Type: new Abstract: Accurate energy demand forecasting is essential for the reliable operation and planning of modern sustainable energy systems. Spatial-temporal graph neu

Reselling “excess AI processing capacity”: first SPCX, now META. In a rational world these moves would be seen as signs that we have *alre…

Reselling “excess AI processing capacity”: first SPCX, now META. In a rational world these moves would be seen as signs that we have *already* started to overbuild. Boom. $META is developing a cloud s

safetygary-marcus--x

Resolving superposition in AI for interpretability and cross-modal alignment in patient-neuronal images

arXiv:2606.31394v1 Announce Type: cross Abstract: Artificial intelligence is transforming our capability to solve biological challenges. In dimensionality bottleneck regimes exacerbated by high-dimens

Rethinking On-policy Optimization for Query Augmentation

arXiv:2510.17139v3 Announce Type: replace Abstract: Recent advances in large language models (LLMs) have led to a surge of interest in query augmentation for information retrieval (IR). Two main appro

safetyarxiv-cs-cl

Revealing Safety-Critical Scenarios for UTM via Transformer

arXiv:2606.31114v1 Announce Type: new Abstract: Unmanned Traffic Management (UTM) systems are cloud-based platforms designed to manage and coordinate multiple aerial vehicles remotely. UTM systems are

Revisiting the Volume Hypothesis

arXiv:2606.31282v1 Announce Type: new Abstract: Modern deep neural networks often contain far more parameters than needed to fit their training data, yet they achieve impressive generalization. A comm

Revocable Learned State via Process Sidecars

arXiv:2606.30788v1 Announce Type: cross Abstract: Language models are often adapted in stages: a public skill phase, a private memory phase, and a later safety phase that learns to refuse outputs tied

safetyarxiv-cs-cl

Reward Redistribution for CVaR MDPs using a Bellman Operator on L-infinity

arXiv:2602.03778v2 Announce Type: replace-cross Abstract: Tail-end risk measures such as static conditional value-at-risk (CVaR) are used in safety-critical applications to prevent rare, yet catastrop

Rhythm-Structured Predictive Learning for Remote Photoplethysmography

arXiv:2606.31736v1 Announce Type: new Abstract: Remote photoplethysmography (rPPG) estimates physiological signals from facial videos by analyzing subtle pulse induced skin color variations. Despite r

Robust Text Watermarking for Large Language Models via Dual Semantic Embeddings

arXiv:2606.31602v1 Announce Type: new Abstract: This work presents Dual-Embedding Watermarking (DEW), a semantic watermarking scheme for large language models (LLMs) that leverages contextual and toke

safetyarxiv-cs-cl

Robustness of Robotic Manipulation: Foundations and Frontiers

arXiv:2606.31494v1 Announce Type: cross Abstract: Humans and animals exhibit remarkable robustness in physical manipulation, yet robots remain far behind. Progress toward human-level manipulation robu

Safe Online Learning via Smooth Safety-Structured Policy Composition

arXiv:2606.31320v1 Announce Type: new Abstract: Safe online reinforcement learning requires policies to respect safety constraints while maintaining smooth optimization dynamics. Existing approaches t

Sampling-Based Coordination-Informed Multi-Objective Multi-Robot Reinforcement Learning

arXiv:2606.30893v1 Announce Type: new Abstract: Multi-robot systems must simultaneously optimize competing objectives while maintaining coordinated behavior. Existing multi-agent reinforcement learnin

Seeing Is Not Sharing: Some Vision-Language Models Overestimate Common Ground in Asymmetric Dialogue

arXiv:2606.31719v1 Announce Type: cross Abstract: In collaborative dialogue, shared perception does not guarantee shared interpretation. Mutual understanding must be established through interaction. W

ShardNet: Training Neural Controllers with Hard, Non-Convex Constraints

arXiv:2606.30935v1 Announce Type: cross Abstract: While neural network control policies are powerful, their deployment on safety critical systems depends on ensuring that they obey strict constraints.

Size Doesn't Matter: Cosine-Scored Sparse Autoencoders

arXiv:2606.15054v2 Announce Type: replace Abstract: Sparse autoencoders (SAEs) detect features via inner product, so a feature's activation scales with both its directional alignment and the input's n

Smart charging of large fleets of Electric Vehicles: Independent Multi-Agent Reinforcement Learning approaches

arXiv:2606.31347v1 Announce Type: new Abstract: The electrification of transportation through electric vehicles introduces new challenges for power grid management, such as increased peak demand, volt

SpectralSplats: Robust Differentiable Tracking via Spectral Moment Supervision

arXiv:2603.24036v2 Announce Type: replace Abstract: 3D Gaussian Splatting (3DGS) enables real-time, photorealistic novel view synthesis, making it a highly attractive representation for model-based vi

Stabilization Learning: A Paradigm Transition Bridging Control Theory and Machine Learning

arXiv:2606.31562v1 Announce Type: new Abstract: Stabilization learning is an interdisciplinary paradigm that bridges control theory and machine learning. Its core idea is to enable systems to adjust t

Stealthy Multi-Task Adversarial Attacks

arXiv:2411.17936v2 Announce Type: replace-cross Abstract: Deep neural networks are highly vulnerable to adversarial perturbations, raising serious safety concerns in the real-world systems. While prio

Structural Preservation and the Logical Expressiveness of Graph Neural Networks

arXiv:2606.17882v2 Announce Type: replace Abstract: Bridges between graph neural networks (GNNs) and logical formalisms have been established by fixing architectural choices, such as the types of aggr

SyncCache: Exploiting Asymmetric Dynamics for Fast Audio-Driven Portrait Animation

arXiv:2606.30849v1 Announce Type: new Abstract: Diffusion Transformers (DiTs) have significantly advanced audio-driven portrait animation, but their high computational cost leads to substantial infere

TactX: Learning Shared Tactile Representations Across Diverse Sensors

arXiv:2606.31236v1 Announce Type: new Abstract: Tactile sensors provide critical information for contact-rich manipulation, yet tactile representations and policies remain tightly coupled to each spec

TDGT: A Tabular Data Generation Toolkit supporting adaptive GPU-accelerated Bayesian mixture models, diffusion-based models, and latent-space generative modeling

arXiv:2606.31268v1 Announce Type: cross Abstract: The growing demand for privacy-preserving data sharing has positioned synthetic data generation as a critical component of responsible AI workflows. D

Test-Time Verification for Text-to-SQL via Outcome Reward Models

arXiv:2606.30851v1 Announce Type: cross Abstract: Improving the reliability of large language models (LLMs) at inference time is a central challenge in structured reasoning tasks such as Text-to-SQL.

That day when @elonmusk starts to sound a bit like my 2023 TED Talk, which called for “a global, nonprofit organization to regulate the tech…

That day when @elonmusk starts to sound a bit like my 2023 TED Talk, which called for “a global, nonprofit organization to regulate the tech for the sake of democracy and our collective future.” Good

safetygary-marcus--x

The Past Is Prologue: A Plug-in Controller for Selective Updates in Sequentially Evolving LLM Memory

arXiv:2606.31121v1 Announce Type: new Abstract: Sequentially evolving LLM memory enables agents to reuse past experience, but existing systems usually deploy each locally generated memory update witho

Token-Sparse Medical Multimodal Reasoning via Dual-Stream Reinforcement Learning

arXiv:2606.31599v1 Announce Type: cross Abstract: Vision-language models (VLMs) combining reinforcement learning (RL) ignite remarkable progress in multimodal reasoning, yet still struggle with medica

TORA: Topological Representation Alignment for 3D Shape Assembly

arXiv:2604.04050v2 Announce Type: replace Abstract: Flow-matching methods for 3D shape assembly learn point-wise velocity fields that transport parts toward assembled configurations, yet they receive

Toxicity Assessment in Preclinical Histopathology via Class-Aware Mahalanobis Distance for Known and Novel Anomalies

arXiv:2602.02124v2 Announce Type: replace-cross Abstract: Drug-induced toxicity is a leading cause of preclinical and early-clinical failure, making early detection critical. Histopathology is the gol

TraCeS: Learning Per-Timestep Constraint-Violation Credit from Sparse Trajectory-Level Labels

arXiv:2504.12557v3 Announce Type: replace-cross Abstract: Ensuring safe behavior in reinforcement learning (RL) is challenging when safety constraints are implicit and cannot be densely measured. In m

Training Therapeutic Judges and Multi-Agent Systems for Human-Aligned Mental Health Support

arXiv:2606.30887v1 Announce Type: cross Abstract: Large language models show promise for mental health support, yet therapeutic quality improves only when evaluation functions as an actionable control

Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation

arXiv:2606.31184v1 Announce Type: cross Abstract: Adaptive experiments for average treatment effects (ATE) require randomized allocations balancing valid inference with statistical efficiency. The ora

TreeAgent: A Generalizable Multi-Agent Framework for Automated Bias Labeling in Forestry via Compiled Expert Rules and Vision-Language Models

arXiv:2606.31976v1 Announce Type: new Abstract: Human-labeled data are widely used as reference annotations in ML, despite known variability across annotators in many expert-driven domains. In additio

TRIAGE: Role-Typed Credit Assignment for Agentic Reinforcement Learning

arXiv:2606.32017v1 Announce Type: cross Abstract: Agentic reinforcement learning requires assigning credit to environment-facing actions such as searches, clicks, edits, navigation commands, and objec