Search: “safety”

ProPACT: A Proactive AI-Driven Adaptive Collaborative Tutor for Pair Programming

arXiv:2605.02703v1 Announce Type: cross Abstract: Effective pair programming depends on coordination of attention, cognitive effort, and joint regulation over time, yet most adaptive learning systems

ProtoFair: Fair Self-Supervised Contrastive Learning via Pseudo-Counterfactual Pairs

DGX agent

arXiv:2605.01971v1 Announce Type: new Abstract: Self-supervised learning methods learn high-quality visual representations, yet recent studies show that these representations often capture demographic

safetyarxiv-cs-cv

5 May 2026

RA-CMF: Region-Adaptive Conditional MeanFlow for CT Image Reconstruction

arXiv:2605.00901v1 Announce Type: new Abstract: The use of CT imaging is important for screening, diagnosis, therapy planning, and prognosis of lung cancers. Unfortunately, due to differences in imagi

Rationality Measurement and Theory for Reinforcement Learning Agents

arXiv:2602.04737v2 Announce Type: replace Abstract: This paper proposes a suite of rationality measures and associated theory for reinforcement learning agents, a property increasingly critical yet ra

RECAP: Transparent Inference-Time Emotion Alignment for Medical Dialogue Systems

arXiv:2509.10746v3 Announce Type: replace Abstract: Large language models in healthcare often produce emotionally flat or opaque responses, failing to provide the transparent reasoning required for cl

safetyarxiv-cs-cl

Registration-Free Learnable Multi-View Capture of Faces in Dense Semantic Correspondence

arXiv:2605.01450v1 Announce Type: new Abstract: Recent frameworks like ToFu and TEMPEH provide an automated alternative to classical registration pipelines by predicting 3D meshes in dense semantic co

Remote Action Generation: Remote Control with Minimal Communication

arXiv:2605.01833v1 Announce Type: cross Abstract: We address the challenge of remote control where one or more actors, lacking direct reward access, are steered by a controller over a communication-co

Representation learning from OCT images

arXiv:2605.02589v1 Announce Type: new Abstract: Optical Coherence Tomography (OCT) has become one of the most used imaging modality in ophthalmology. It provides high-resolution, non-invasive visualiz

Research on Vision-Language Question Answering Models for Industrial Robots

arXiv:2605.01483v1 Announce Type: new Abstract: A hierarchical cross-modal fusion model is proposed for vision-language question answering (VLQA) in industrial robotics, targeting the challenges of se

Rethinking Model Selection in VLM Through the Lens of Gromov-Wasserstein Distance

arXiv:2605.01325v1 Announce Type: new Abstract: Vision-Language Models (VLMs) have enhanced traditional LLMs with visual capabilities through the integration of vision encoders. While recent works hav

Retrieval-Guided Generation for Safer Histopathology Image Captioning

arXiv:2605.00893v1 Announce Type: new Abstract: Generative vision-language models can produce fluent medical image captions but remain prone to hallucination, over-specific diagnostic claims, and fact

Rhythm: Learning Interactive Whole-Body Control for Dual Humanoids

arXiv:2603.02856v2 Announce Type: replace Abstract: Realizing interactive whole-body control for multi-humanoid systems is critical for unlocking complex collaborative capabilities in shared environme

safetyarxiv-cs-ro

Robo3R: Enhancing Robotic Manipulation with Accurate Feed-Forward 3D Reconstruction

arXiv:2602.10101v2 Announce Type: replace Abstract: 3D spatial perception is fundamental to generalizable robotic manipulation, yet obtaining reliable, high-quality 3D geometry remains challenging. De

safetyarxiv-cs-ro

Semantic-Contact Fields for Category-Level Generalizable Tactile Tool Manipulation

arXiv:2602.13833v2 Announce Type: replace Abstract: Generalizing tool manipulation requires both semantic planning and precise physical control. Modern generalist robot policies, such as Vision-Langua

safetyarxiv-cs-ro

Sentra-Guard: A Real-Time Multilingual Defense Against Adversarial LLM Prompts

arXiv:2510.22628v2 Announce Type: replace-cross Abstract: This paper presents a real-time modular defense system named Sentra-Guard. The system detects and mitigates jailbreak and prompt injection att

Separation Assurance between Heterogeneous Fleets of Small Unmanned Aerial Systems via Multi-Agent Reinforcement Learning

arXiv:2605.01041v1 Announce Type: cross Abstract: In the envisioned future dense urban airspace, multiple companies will operate heterogeneous fleets of small unmanned aerial systems (sUASs), where ea

SIFT-VTON: Geometric Correspondence Supervision on Cross-Attention for Virtual Try-On

arXiv:2605.01296v1 Announce Type: new Abstract: Diffusion-based virtual try-on methods achieve photorealistic synthesis through cross-attention mechanisms that transfer garment features to target body

SignVerse-2M: A Two-Million-Clip Pose-Native Universe of 25+ Sign Languages

arXiv:2605.01720v1 Announce Type: cross Abstract: Existing large-scale sign language resources typically provide supervision only at the level of raw video-text alignment and are often produced in lab

safetyarxiv-cs-cl

Singular Bayesian Neural Networks

arXiv:2602.00387v3 Announce Type: replace-cross Abstract: Bayesian neural networks promise calibrated uncertainty but require O(mn) parameters for standard mean-field Gaussian posteriors. We argue thi

Skills as Verifiable Artifacts: A Trust Schema and a Biconditional Correctness Criterion for Human-in-the-Loop Agent Runtimes

arXiv:2605.00424v1 Announce Type: cross Abstract: Agent skills -- structured packages of instructions, scripts, and references that augment a large language model (LLM) without modifying the model its

So let me make sure I’ve got this correct: >Google gives Anthropic cash to pay Google for compute. >anthropic gives compute away for a loss,…

So let me make sure I’ve got this correct: >Google gives Anthropic cash to pay Google for compute. >anthropic gives compute away for a loss, and this discounted compute draws in large user bases to ca

safetygary-marcus--x

Sonar-GPS Fusion for Seabed Mapping in Turbid Shallow Waters with an Autonomous Surface Vehicle

arXiv:2605.01949v1 Announce Type: cross Abstract: Accurate seabed mapping is essential for habitat monitoring and infrastructure inspection. In turbid, shallow coastal waters, such as shellfish aquacu

Sources: the WH is mulling EOs to address security risks from advanced AI, including barring companies from 'interfering' with the government's use of models (Politico)

Politico: Sources: the WH is mulling EOs to address security risks from advanced AI, including barring companies from “interfering” with the government's use of models — A White House spokesperson sai

safetytechmeme

SpectraDINO: Bridging the Spectral Gap in Vision Foundation Models via Lightweight Adapters

arXiv:2605.02258v1 Announce Type: new Abstract: Vision Foundation Models (VFMs) pretrained on large-scale RGB data have demonstrated remarkable representation quality, yet their applicability to multi

SRA: Span Representation Alignment for Large Language Model Distillation

arXiv:2605.01205v1 Announce Type: new Abstract: Cross-Tokenizer Knowledge Distillation (CTKD) enables knowledge transfer between a large language model and a smaller student, even when they employ dif

safetyarxiv-cs-cl

Model Releases

SRTJ: Self-Evolving Rule-Driven Training-Free LLM Jailbreaking

model-releasesarxiv-cs-cl

arXiv:2605.00974v1 Announce Type: cross Abstract: LLMs are increasingly equipped with safety alignment mechanisms, yet recent studies demonstrate that they remain vulnerable to jailbreaking attacks th

StableMind: Source-Free Cross-Subject fMRI Decoding with Regularized Adaptation

arXiv:2605.02586v1 Announce Type: new Abstract: Existing cross-subject fMRI decoding methods typically train a model on multiple scanned subjects and then adapt it to a new subject using substantial p

STABLEVAL: Disagreement-Aware and Stable Evaluation of AI Systems

arXiv:2605.02122v1 Announce Type: new Abstract: Human evaluation remains the primary standard for assessing modern AI systems, yet annotator disagreement, bias, and variability make system rankings fr

SwiftPie: Lightning-fast Subject-driven Image Personalization via One step Diffusion

arXiv:2605.01510v1 Announce Type: new Abstract: Diffusion models have achieved remarkable success in high-quality image synthesis, sparking interest in image-guided generation tasks such as subject-dr

SynPAIN: A Synthetic Dataset of Pain and Non-Pain Facial Expressions

arXiv:2507.19673v3 Announce Type: replace Abstract: Accurate pain assessment in patients with limited ability to communicate, such as older adults with severe dementia, represents a critical healthcar

Task-Related Token Compression in Multimodal Large Language Models from an Explainability Perspective

arXiv:2506.01097v2 Announce Type: replace Abstract: Existing Multimodal Large Language Models (MLLMs) process a large number of visual tokens, leading to significant computational costs and inefficien

Thank you @UN Secretary General @antonioguterres for an excellent discussion on the current state of AI capabilities, and the urgent need fo…

Thank you @UN Secretary General @antonioguterres for an excellent discussion on the current state of AI capabilities, and the urgent need for robust global governance. International collaboration is c

safetyyoshua-bengio--x

The Case for ESM3 as a General-Purpose AI Model with Systemic Risk Under the EU AI Act

arXiv:2605.01611v1 Announce Type: cross Abstract: Due to ambiguity in the wording of the EU AI Act, we examine the question of to what extent frontier biological foundation models such as ESM3 are sub

The Geometric Inductive Bias of Grokking: Bypassing Phase Transitions via Architectural Topology

arXiv:2603.05228v3 Announce Type: replace Abstract: Mechanistic interpretability typically relies on post-hoc analysis of trained networks. We instead adopt an interventional approach: testing hypothe

The Geometric Mechanics of Contrastive Representation Learning: Alignment Potentials, Entropic Dispersion, and Cross-modal Divergence

arXiv:2601.19597v3 Announce Type: replace Abstract: While InfoNCE underlies modern contrastive learning, its geometric mechanisms remain under-characterized beyond the canonical alignment--uniformity

The Model Knows, the Decoder Finds: Future Value Guided Particle Power Sampling

arXiv:2605.02427v1 Announce Type: cross Abstract: A recurring pattern in 'reasoning without training' is that base LLMs already assign non-trivial probability mass to correct multi-step solutions; the

The Partial Testimony of Logs: Evaluation of Language Model Generation under Confounded Model Choice

arXiv:2605.01311v1 Announce Type: new Abstract: Offline evaluation of language models from usage logs is biased when model choice is confounded: the same user-side factors that influence which model i

The Pragmatic Frames of Spurious Correlations in Machine Learning: Interpreting How and Why They Matter

arXiv:2411.04696v5 Announce Type: replace Abstract: Learning correlations from data forms the foundation of today's machine learning (ML) and artificial intelligence research. While contemporary metho

Today, the CEO of @coinbase announced a 14% headcount reduction. Let's talk about why. He cites AI as a core reason for the shift, alongside…

Today, the CEO of @coinbase announced a 14% headcount reduction. Let's talk about why. He cites AI as a core reason for the shift, alongside market dynamics - though we don't know whether each of thos

safetyallie-k--miller--x

Too early to tell how the trial ends, but if Elon does win, the takeaway might be in part that Greg Brockman made two immense blunders: 1. W…

Too early to tell how the trial ends, but if Elon does win, the takeaway might be in part that Greg Brockman made two immense blunders: 1. Writing down all that stuff in his diary 2. Blowing off Elon’

safetygary-marcus--x

Topological Neural Tangent Kernel

arXiv:2605.01110v1 Announce Type: new Abstract: Graph neural tangent kernels give a principled infinite-width theory for graph neural networks, but inherit a basic limitation of graph models: they see

Toward a Scientific Discovery Engine for Weather and Climate Data: A Visual Analytics Workbench for Embedding-Based Exploration

arXiv:2605.00972v1 Announce Type: cross Abstract: Earth system science is producing increasingly large, high-dimensional datasets from physics based Earth system models to AI-based weather and climate

Towards Efficient and Expressive Offline RL via Flow-Anchored Noise-conditioned Q-Learning

arXiv:2605.01663v1 Announce Type: new Abstract: We propose Flow-Anchored Noise-conditioned Q-Learning (FAN), a highly efficient and high-performing offline reinforcement learning (RL) algorithm. Recen

Towards Improving Speaker Distance Estimation through Generative Impulse Response Augmentation

arXiv:2605.00721v1 Announce Type: cross Abstract: The Room Acoustics and Speaker Distance Estimation (SDE) Challenge at ICASSP 2025 explores the effectiveness of augmented room impulse response (RIR)

Training Non-Differentiable Networks via Optimal Transport

arXiv:2605.01928v1 Announce Type: new Abstract: Neural networks increasingly embed non-differentiable components (spiking neurons, quantized layers, discrete routing, blackbox simulators, etc.) where

TRAP: Tail-aware Ranking Attack for World-Model Planning

arXiv:2605.01950v1 Announce Type: new Abstract: World models enable long-horizon planning by internally generating and evaluating imagined trajectories, making them a promising foundation for generali

TUR-DPO: Topology- and Uncertainty-Aware Direct Preference Optimization

arXiv:2605.00224v1 Announce Type: new Abstract: Aligning large language models (LLMs) with human preferences is commonly done via reinforcement learning from human feedback (RLHF) with Proximal Policy

Ultrasound Vision-Language Alignment via Contrastive Learning

arXiv:2605.02126v1 Announce Type: new Abstract: Ultrasound foundation models have achieved strong performance on structured prediction tasks but remain exclusively vision-based, limiting zero-shot and