Search: “safety”

MARS: Margin and Semantic-Aware Data Augmentation for Reward Modeling

arXiv:2602.17658v2 Announce Type: replace-cross Abstract: Reward modeling is central to alignment pipelines such as RLHF, RLAIF, and PPO-based policy optimization, yet its reliability is constrained b

MATO: Multi-objective Personalized Alignment with Test-time Optimization for Large Language Models

DGX agent

arXiv:2605.25342v1 Announce Type: new Abstract: Aligning large language models (LLMs) with diverse and multifaceted user preferences is a fundamental challenge in personalized AI systems. Existing mul

safetyarxiv-cs-cl

26 May 2026

MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents

arXiv:2602.02474v2 Announce Type: replace-cross Abstract: Most Large Language Model (LLM) agent memory systems rely on a small set of static, hand-designed operations for extracting memory. These fixe

Metacognition Should Be the Scientific Framework for Bounded and Effective Self-Governance in Generative AI

arXiv:2605.23981v1 Announce Type: cross Abstract: Generative AI research increasingly confronts a shared problem: systems must sustain yet govern their own generative activity when uncertainty is high

Micro-Swarm Locomotion Optimization in Dynamic Flow using Multi-Objective Multi-Agent Reinforcement Learning

arXiv:2605.25025v1 Announce Type: new Abstract: Coordinating micro-robotic swarms in physiologically realistic, time-dependent fluid environments remains an unsolved challenge for biomedical and envir

MMUEChange: A Generalized LLM Agent Framework for Intelligent Multi-Modal Urban Environment Change Analysis

arXiv:2601.05483v2 Announce Type: replace Abstract: Understanding urban environment change is essential for sustainable development. However, current approaches, particularly remote sensing change det

Motion-Compensated Weight Compression

arXiv:2605.24754v1 Announce Type: cross Abstract: Neural network weights are increasingly a bottleneck for deployment, yet most compression pipelines treat layers independently and overlook cross-laye

MuGen: Multi-Skill Generative Locomotion Controller for Humanoid Robots

arXiv:2605.24592v1 Announce Type: new Abstract: This paper presents MuGen, a data-driven framework for learning and deploying multi-skill locomotion on humanoid robots. MuGen enables a robot to perfor

Multi-Agent Coordination Adaptation via Structure-Guided Orchestration

arXiv:2605.25746v1 Announce Type: cross Abstract: As large language model (LLM)-based multi-agent systems scale to handle increasingly complex tasks, balancing structural stability and dynamic adaptab

Multi-Agent Systems are Mixtures of Experts: Who Becomes an Influencer?

arXiv:2605.25929v1 Announce Type: cross Abstract: The effectiveness of multi-agent LLM deliberation depends not only on the agents' individual predictions, but also on how they communicate and collabo

Multi-Alignment Contrastive Learning for Enzyme--Reaction Retrieval

arXiv:2512.08508v2 Announce Type: replace-cross Abstract: Identifying enzymes that catalyze target biochemical reactions is a key step in computational enzyme discovery and biocatalyst design. Recent

Multi-Modal Cross-Domain Alignment Network for Video Moment Retrieval

arXiv:2209.11572v3 Announce Type: replace-cross Abstract: As an increasingly popular task in multimedia information retrieval, video moment retrieval (VMR) aims to localize the target moment from an u

Multi-Objective Learning for Diffusion Models: A Statistical Theory under Semi-Supervised Learning

arXiv:2605.25210v1 Announce Type: cross Abstract: Diffusion models are increasingly used as powerful conditional generators, yet real deployments often involve multiple target distributions arising fr

Multi-view Consistent 3D Gaussian Head Avatars 'without' Multi-view Generation

arXiv:2605.25220v1 Announce Type: cross Abstract: High-fidelity 3D Gaussian head avatar generation is critical for applications such as AR/VR, telepresence, and digital humans. Existing methods depend

Multicalibration Boosting: Theory, Convergence, and Transferability

arXiv:2605.24364v1 Announce Type: cross Abstract: Multicalibration extends classical calibration by requiring predictions to be unbiased over a rich collection of functions, encompassing both predicti

Multimodal Alignment and Preference Optimization for Zero-Shot Conditional RNA Generation

arXiv:2605.23961v1 Announce Type: cross Abstract: The design of RNA molecules that interact with specific proteins is a critical challenge in experimental and computational biology. Despite recent pro

Multimodal Functional Maximum Correlation for Emotion Recognition

arXiv:2512.23076v2 Announce Type: replace-cross Abstract: Emotional states manifest as coordinated yet heterogeneous physiological responses across central and autonomic systems, posing a fundamental

MultiPhishGuard: An Explainable and Adaptive Multi-Agent LLM System for Phishing Email Detection

arXiv:2505.23803v2 Announce Type: replace-cross Abstract: Phishing email detection faces significant challenges due to evolving adversarial tactics and heterogeneous attack patterns. Traditional appro

Music Transcription with (Almost) No Supervision

arXiv:2605.24193v1 Announce Type: cross Abstract: Competitive music transcription models require large amounts of paired audio-score data, which is scarce due to collection costs, alignment difficulty

NeuralTouch: Neural Descriptors for Precise Sim-to-Real Tactile Robot Control

arXiv:2510.20390v2 Announce Type: replace Abstract: Grasping accuracy is a critical prerequisite for precise object manipulation, often requiring careful alignment between the robot hand and object. N

Neuro-Inspired Inverse Learning for Planning and Control

arXiv:2605.24152v1 Announce Type: new Abstract: We present a neuro-inspired framework for embodied planning and control. Building on three principles that enable fast and highly effective goal-directe

Not All Transitions Matter: Evidence from PPO

arXiv:2605.24071v1 Announce Type: cross Abstract: Training a reinforcement learning agent on-policy means collecting fresh experience at every update, and that experience comes with a hidden problem.

Not only where, But when: Temporal Scheduling for RLVR

arXiv:2605.25381v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) has become a core technique for post-training of Large Language Models (LLMs). While policy optimi

OASIS: Observation-Action Space Alignment via SE(3) Trajectory Prediction for Robotic Manipulation

arXiv:2605.25829v1 Announce Type: cross Abstract: Recent vision-language-action (VLA) models and world action models (WAMs) advance robotic manipulation by enriching intermediate representations with

oh. my. god. could this word cloud diagram be … conscious?

Gary Marcus, a prominent AI researcher and critic, questions whether a word cloud diagram could possess consciousness, likely engaging in ironic commentary on overclaimed AI capabilities or consciousn

safetygary-marcus--x

OmniSapiens: A Foundation Model for Social Behavior Processing via Heterogeneity-Aware Relative Policy Optimization

arXiv:2602.10635v2 Announce Type: replace Abstract: Socially intelligent AI systems must entail reasoning across diverse human behavioral tasks, and generalization to new contexts. However, AI has yet

On Reliability of Efficient Membership Inference Vulnerability Evaluation

arXiv:2605.25819v1 Announce Type: new Abstract: Membership inference attacks (MIAs) are popular methods for empirically assessing the leakage of sensitive information in the training data through mode

On the Benefits of Free Exploration for Regret Minimization in Multi-Armed Bandits

arXiv:2605.25789v1 Announce Type: cross Abstract: We study a stochastic multi-armed bandit problem where an agent is granted a free exploration budget before regret accumulates, a setting not captured

One-Step Bellman Alignment Enables Provably Efficient Transfer in Online RL

arXiv:2601.21924v2 Announce Type: replace Abstract: We study online transfer reinforcement learning (RL) in episodic Markov decision processes, where experience from related source tasks is available

openai truly tanked a golden reputation, and squandered its lead, yet its board stands by its CEO, and looks for a band aid.

openai truly tanked a golden reputation, and squandered its lead, yet its board stands by its CEO, and looks for a band aid. OpenAI has a PR challenge, and although it's spoken to several candidates f

safetygary-marcus--x

Optimizing Token Choice for Code Watermarking: An RL Approach

arXiv:2508.11925v3 Announce Type: replace-cross Abstract: Protecting intellectual property on LLM-generated code necessitates effective watermarking systems that can operate within code's highly struc

PageLLM: A Multi-Grained Reward Framework for Whole-Page Optimization with Large Language Models

arXiv:2506.09084v2 Announce Type: replace-cross Abstract: Whole-page optimization (WPO) decides how search and recommendation results are surfaced to users, and large language models (LLMs) open a new

“pass me the crack pipe” @edels0n on SpaceX’s ludicrous valuation:

Gary Marcus critiques SpaceX's valuation as unreasonably inflated, using hyperbolic language to suggest the company's market assessment lacks rational justification. The post likely discusses concerns

safetygary-marcus--x

PathWise: Planning through World Model for Automated Heuristic Design via Self-Evolving LLMs

arXiv:2601.20539v3 Announce Type: replace Abstract: Large Language Models (LLMs) have enabled automated heuristic design (AHD) for combinatorial optimization problems (COPs), but existing frameworks'

Peak-Then-Collapse and the Four Interface Channels of Knowledge-Graph Tool Use

arXiv:2605.26037v1 Announce Type: new Abstract: We test the standard RLVR tool-use recipe -- GRPO on Qwen2.5-7B-Instruct -- on a deliberately minimal knowledge-graph tool API: four Freebase navigation

PID-Guided Partial Alignment for Multimodal Decentralized Federated Learning

arXiv:2601.10012v2 Announce Type: replace Abstract: Multimodal decentralized federated learning (DFL) must support collaboration among agents that hold different modality subsets and often different m

PILOT: Policy-Informed Learned Optimization for Adaptive Deep Network Training

arXiv:2605.24570v1 Announce Type: cross Abstract: Despite the central role of optimization in deep learning, most optimizers rely on update structures whose functional form is fixed before training be

PolyGnosis 2.0: Enhancing LLM Reasoning via Agentic Harness Engineering for Polymarket and OSINT Insight Extraction

arXiv:2605.25958v1 Announce Type: new Abstract: This paper introduces PolyGnosis 2.0, a pioneering multi-agent architecture designed to extract predictive intelligence by synthesizing Polymarket anoma

Polynomial Context-Truncation Sensitivity in Autoregressive Language Models: Sequential Wyner-Ziv Bounds for KV Cache Compression

arXiv:2605.25085v1 Announce Type: cross Abstract: We study the rate-distortion limits of online KV cache compression in autoregressive language models, formulating it as sequential Wyner-Ziv source co

PrivFusion: A Privacy-preserving Multi-Agent Framework for Harmonizing Distributed Datasets

arXiv:2605.24249v1 Announce Type: new Abstract: The growing availability of clinical data has increased the use of machine learning, yet centralized data aggregation is often infeasible for sensitive

ProActor: Timing-Aware Reinforcement Learning for Proactive Task Scheduling Agents

arXiv:2605.24900v1 Announce Type: new Abstract: Proactive task-oriented agents must autonomously anticipate user needs, identify actionable opportunities, and trigger software actions at appropriate m

Quantitative Evaluation of the Severity of Posttraumatic Stress Disorder through Transfer Learning from Specific Phobia Data

arXiv:2605.25933v1 Announce Type: cross Abstract: Posttraumatic stress disorder (PTSD) is a prevalent and debilitating mental health condition with significant personal and societal impacts. Current c

Quantum Frog: Emergent Cooperation and Difficulty Scaling in a Quantized-Time Cooperative Game

arXiv:2605.23930v1 Announce Type: new Abstract: We introduce Quantum Frog, a two-player cooperative game built on a novel quantized-time mechanic in which the environment advances only when a player a

Reading, Not Thinking: Understanding and Bridging the Modality Gap When Text Becomes Pixels in Multimodal LLMs

arXiv:2603.09095v2 Announce Type: replace Abstract: Multimodal large language models (MLLMs) can process text presented as images, yet they often perform worse than when the same content is provided a

Reasoning as an Attack Surface: Adaptive Evolutionary CoT Jailbreaks for LLMs

arXiv:2605.24497v1 Announce Type: new Abstract: Large Reasoning Models (LRMs) have demonstrated remarkable capabilities in reasoning and generation tasks and are increasingly deployed in real-world ap

RecGOAT: Graph Optimal Adaptive Transport for LLM-Enhanced Multimodal Recommendation with Dual Semantic Alignment

arXiv:2602.00682v2 Announce Type: replace-cross Abstract: Integrating large language model (LLM) representations into multimodal recommendation has shown promise, yet a fundamental challenge remains l

Refined Analysis of Entropy-Regularized Actor-Critic

arXiv:2605.24357v1 Announce Type: new Abstract: In this paper, we study the role of the critic in actor--critic for entropy-regularized, finite, discounted environments. We establish that, when the cr

Reinforcement Learning from Denoising Feedback

arXiv:2605.25638v1 Announce Type: new Abstract: Policy loss estimation remains a fundamental and long-standing challenge in reinforcement learning (RL) for diffusion language models (dLLMs). We introd