Search: “safety”

Fourier Features Let Agents Learn High Precision Policies with Imitation Learning

arXiv:2606.12334v1 Announce Type: new Abstract: High-precision robotic manipulation requires fine-grained spatial reasoning that is often difficult to achieve with RGB-only policies due to depth ambig

FreeBridge: Variational Schrodinger Bridges for Cellular Transition Dynamics

DGX agent

arXiv:2606.11286v1 Announce Type: cross Abstract: High-content imaging assays quantify cellular responses to chemical and genetic perturbations, yet continuous trajectories of individual cells are uno

safetyarxiv-cs-ai

11 Jun 2026

FreqKD: Frequency-Decoupled Cross-Modal Knowledge Distillation for Infrared Object Detection

arXiv:2606.11572v1 Announce Type: new Abstract: Transfer learning from large-scale RGB foundation models to infrared (IR) imagery through knowledge distillation (KD) remains challenging due to fundame

From Architecture to Output: Structural Origins of Hallucination in Large Language Models and the Amplifying Role of Data

arXiv:2606.07537v1 Announce Type: cross Abstract: Large language models hallucinate--producing fluent, confident, factually wrong outputs--with a consistency that persists across generations and scale

From Awareness to Action: Understanding and Overcoming the Research-Practice Gap in Algorithmic Fairness for Public Health

arXiv:2606.11214v1 Announce Type: cross Abstract: Algorithmic fairness is essential for responsible ML-driven public health research, yet its practical implementation remains limited. To investigate t

From Correspondence to Actions: Human-Like Multi-Image Spatial Reasoning in Multi-modal Large Language Models

arXiv:2602.08735v3 Announce Type: replace Abstract: While multimodal large language models (MLLMs) have made substantial progress in single-image spatial reasoning, multi-image spatial reasoning, whic

From Explicit Elements to Implicit Intent: A Predefined Library for Auditable Behavioral Inference

arXiv:2606.11207v1 Announce Type: new Abstract: We present SemantiClean, a modular framework for extracting structured semantic signals from e-commerce session data and driving pluggable inference tar

Generalizing Beyond Suboptimality: Offline Reinforcement Learning Learns Effective Scheduling through Random Solutions

arXiv:2509.10303v2 Announce Type: replace-cross Abstract: Online reinforcement learning (RL) approaches have demonstrated strong performance on Job Shop Scheduling (JSP) and Flexible JSP (FJSP) proble

Geometric bias in eigenspace perturbation under random heterogeneous noise

arXiv:2606.11263v1 Announce Type: cross Abstract: Spectral methods rely fundamentally on the stability of principal eigenspaces under random perturbations. Classically, this stability is quantified by

GLACIER: A Multimodal Student-Teacher Foundation Model for Molecular Property Prediction

arXiv:2606.11382v1 Announce Type: new Abstract: Deep learning models facilitate the discovery of molecules with tailored properties among billions of candidate compounds. However, the computational bu

GPO: Learning from Critical Steps to Improve LLM Reasoning

arXiv:2509.16456v3 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used in various domains, showing impressive potential on different tasks. Recently, reasoning LLMs hav

HERO: Hindsight-Enhanced Reflection from Environment Observations for Agentic Self-Distillation

arXiv:2606.11559v1 Announce Type: new Abstract: Reinforcement learning typically improves multi-turn agent capabilities through the terminal outcome of the trajectories, which makes it difficult to de

Hey Chat, Can You Teach Me? Structuring Socratic Dialogue for Human Learning in the Wild

arXiv:2606.11744v1 Announce Type: cross Abstract: Large language models are now widely used for everyday learning, but the underlying interactions are typically unstructured chats rather than followin

Hot take on a bad day for OpenAI: WeWork’s IPO fell apart. I’ve been calling OpenAI the (possible) WeWork of AI for a long time. I will not …

Hot take on a bad day for OpenAI: WeWork’s IPO fell apart. I’ve been calling OpenAI the (possible) WeWork of AI for a long time. I will not be surprised if their IPO plans apart, too. Their apparent p

https://open.substack.com/pub/garymarcus/p/maybe-section-230-doesnt-shield-ai?r=8tdk6&utm_campaign=post&utm_medium=web&showWelcomeOnShare=tr…

Gary Marcus discusses whether Section 230 of the Communications Decency Act adequately protects AI systems and companies from legal liability, suggesting that existing protections may not fully apply

IAPO: Input Attribution-Aware Policy Optimization for Tool Use in Small Multimodal Agents

arXiv:2606.11652v1 Announce Type: new Abstract: This paper investigates reinforcement learning (RL) methods for improving tool-calling capabilities in multimodal small language model (SLM) agents. Whi

If OpenAI fails, will you be sad?

Gary Marcus poses a philosophical question about emotional attachment to AI companies, likely exploring whether people have genuine emotional stakes in the success or failure of specific AI organizati

Illumination-Robust Camera-Based Heart-Rate Estimation for Physiological Sensing in Robots

arXiv:2606.12378v1 Announce Type: cross Abstract: Physiological awareness is important for service, social, and assistive robots that interact with humans in everyday environments. Remote photoplethys

Implicit Neural Representations of Individual Behavior

arXiv:2606.12200v1 Announce Type: cross Abstract: We study policy representation learning from unlabeled multi-policy behavioral data. Each episode is generated by a fixed policy, but policy labels ar

Improving Generalization and Data Efficiency with Diffusion in Offline Multi-agent RL

arXiv:2307.01472v2 Announce Type: replace Abstract: We present a novel Diffusion Offline Multi-agent Model (DOM2) for offline Multi-Agent Reinforcement Learning (MARL). Different from existing algorit

IntElicit: Eliciting and Assessing Contextualized Creativity via Dialogue Policy Optimization

arXiv:2606.12086v1 Announce Type: new Abstract: Contextualized assessment offers high ecological validity for evaluating creativity but introduces a critical challenge: observed performance may be con

InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning

arXiv:2606.12195v1 Announce Type: new Abstract: Recent progress in foundation models has shifted toward agentic behavior involving multi-step reasoning and tool use. However, open-source efforts large

is that more or less times than Elon said robotaxis would be ubiquitous within a year?

This post by AI researcher Gary Marcus likely compares a prediction or timeline claim about robotaxi deployment to Elon Musk's previous statements about achieving ubiquitous robotaxis within a year, q

ISAP-3D: Identity-Slot Aligned Part-Aware 3D Generation

arXiv:2606.12099v1 Announce Type: new Abstract: Part-aware 3D generation aims to synthesize structured objects with semantically meaningful components, yet often suffers from structural ambiguity due

KinematicRL: A Sim-to-Real Reinforcement Learning Framework For Social Navigation With Kinodynamic Feasibility

arXiv:2606.12042v1 Announce Type: new Abstract: Deep Reinforcement Learning (DRL) has shown promise for social navigation, yet its real-world deployment remains hindered by a persistent sim-to-real ga

safetyarxiv-cs-ro

LAST: Bridging Vision-Language and Action Manifolds via Gromov-Wasserstein Alignment

arXiv:2606.11221v1 Announce Type: new Abstract: We take a Gromov-Wasserstein perspective on Vision-Language-Action (VLA) learning, where the goal is to make the relational geometry of action represent

Latent World Recovery for Multimodal Learning with Missing Modalities

arXiv:2606.12362v1 Announce Type: cross Abstract: We study multimodal learning under missing modalities, with particular motivation from bioscience applications in which heterogeneous modalities are o

Learning Instance-Adaptive Low-Rank Orthogonal Subspaces for Clothes-Changing Person Re-Identification

arXiv:2606.11661v1 Announce Type: new Abstract: Clothes-changing person re-identification (CC-ReID) aims to recognize individuals despite drastic appearance changes caused by clothing variation. While

Learning to Inject: Automated Prompt Injection via Reinforcement Learning

arXiv:2602.05746v2 Announce Type: replace-cross Abstract: Prompt injection is a critical vulnerability in LLM agents, yet the strongest methods still rely on human red-teamers and hand-crafted prompts

Learning What to Say to Your VLA: Mostly Harmless Vision Language Action Model Steering

arXiv:2606.12299v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models provide a natural language interface to robot control, but the mapping from language to behavior is often brittle

LUCID: Learning Embodiment-Agnostic Intent Models from Unstructured Human Videos for Scalable Dexterous Robot Skill Acquisition

arXiv:2606.11628v1 Announce Type: cross Abstract: The most widely-adopted robot learning pipelines today learn skills from robot demonstrations or structured human data, which are expensive to collect

MASK: Multi-Agent Semantic K-Scheduling for Risk-Sensitive 6G Robotics

arXiv:2606.11249v1 Announce Type: cross Abstract: Realizing the vision of 6G connected robotics requires reconciling high-performance collaborative control with the rigid spectral limitations of physi

Maybe Section 230 doesn’t shield AI companies after all. That would be huge. https://open.substack.com/pub/garymarcus/p/maybe-section-230-do…

Maybe Section 230 doesn’t shield AI companies after all. That would be huge. https://open.substack.com/pub/garymarcus/p/maybe-section-230-doesnt-shield-ai?r=8tdk6&utm_campaign=post&utm_medium=web&show

Measuring Semantic Progress in Multi-turn Dialogue via Information Gain

arXiv:2606.12332v1 Announce Type: new Abstract: Evaluating multi-turn dialogue is challenging because quality emerges across turns rather than within individual responses. We focus on a key dimension

safetyarxiv-cs-cl

Mirror Descent Beyond Euclidean Stability: An Exponential Separation in Initialization Sensitivity

arXiv:2606.11431v1 Announce Type: new Abstract: Mirror Descent (MD) extends Gradient Descent (GD) beyond Euclidean geometry and has recently reappeared as a lens for KL-regularized policy optimization

Mitigating Disparate Impact of Differentially Private Learning through Bounded Adaptive Clipping

arXiv:2506.01396v2 Announce Type: replace Abstract: Differential privacy (DP) has become an essential framework for privacy-preserving machine learning. Existing DP learning methods, however, often ha

MLT-Dedup: Efficient Large-Scale Online Video Deduplication via Multi-Level Representations and Spatial-Temporal Matching

arXiv:2606.12215v1 Announce Type: new Abstract: The explosive growth of user-generated video content on online platforms is accompanied by the emergence of numerous near-duplicate videos--videos that

Noise-Guided Transport for Imitation Learning

arXiv:2509.26294v2 Announce Type: replace-cross Abstract: We consider imitation learning in the low-data regime, where only a limited number of expert demonstrations are available. In this setting, me

Offline Diffusion Policy for Multi-User Delay-Constrained Scheduling

arXiv:2501.12942v2 Announce Type: replace Abstract: Effective multi-user delay-constrained scheduling is crucial in various real-world applications, including embodied AI, instant messaging, live stre

Open Materials Generation with Inference-Time Reinforcement Learning

arXiv:2602.00424v2 Announce Type: replace Abstract: Continuous-time generative models for crystalline materials enable inverse materials design by learning to predict stable crystal structures, but in

🚨 OpenAI pondering big price cuts, per WSJ scoop:

OpenAI is reportedly considering significant price reductions for its AI services, according to a Wall Street Journal report shared by AI researcher Gary Marcus. The move would likely reflect competit

PAWS: Preference Learning with Advantage-Weighted Segments

arXiv:2606.11982v1 Announce Type: new Abstract: Preference-based reinforcement learning (PbRL) learns policies from human trajectory-level comparisons, avoiding explicit reward design and expert demon

Plan-and-Verify Video Reward Reasoning with Spatio-Temporal Scene Graph Grounding

arXiv:2606.11838v1 Announce Type: new Abstract: Reward models for text-to-video (T2V) generation guide post-training but often fail at fine-grained semantic alignment. We trace this to two structural

ProcessThinker: Enhancing Multi-modal Large Language Models Reasoning via Rollout-based Process Reward

arXiv:2606.11209v1 Announce Type: cross Abstract: Visual question answering increasingly requires multi-step reasoning. Recent post-training with reinforcement learning under verifiable rewards (RLVR)

Redesign Mixture-of-Experts Routers with Manifold Power Iteration

arXiv:2606.12397v1 Announce Type: cross Abstract: Router is the cornerstone component to the Mixture-of-Experts models. Serving as expert proxies, the rows of the router matrix compute their similarit

Reinforcement Learning Disrupts Gradient-Based Adversarial Optimization

arXiv:2606.12251v1 Announce Type: cross Abstract: Gradient-based adversarial attacks remain a dominant threat to deep neural networks (DNNs), as they exploit gradient information to efficiently optimi

Reinforcement Learning with Action-Triggered Observations

arXiv:2510.02149v2 Announce Type: replace Abstract: We introduce Action-Triggered Sporadically Traceable Markov Decision Processes (ATST-MDPs), a reinforcement learning framework for partial observabi

Reverse Flow Matching: A Unified Framework for Online Reinforcement Learning with Diffusion and Flow Policies

arXiv:2601.08136v2 Announce Type: replace Abstract: Diffusion and flow policies are gaining prominence in online reinforcement learning (RL) due to their expressive power, yet training them efficientl