Search: “safety”

Does Fine-Tuning Undo Activation Steering? Behavioural Recovery Without Weight-Edit Reversal

arXiv:2608.24988v1 Announce Type: new Abstract: Activation steering can be embedded directly into a language model's weights, shaping behaviour without inference-time intervention and offering a way t

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

DualOPSD: Adaptive Privileged Teachers for On-Policy Self-Distillation

DGX agent

arXiv:2608.26019v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) uses a privileged copy of the student model to provide dense supervision without an external teacher. OPSD keeps this

safetyarxiv-cs-lg

27 Aug 2026

Fairness-Aware Test-Time Prompt Tuning

arXiv:2608.25707v1 Announce Type: new Abstract: Vision-language models have displayed remarkable capabilities in multi-modal understanding and are increasingly used in critical applications where econ

FRAME: separating sampling variation from representational cause in medical imaging fairness

arXiv:2608.25981v1 Announce Type: new Abstract: Subgroup performance differences are the standard evidence for fairness bias in medical imaging, and the usual response removes the demographic informat

From National Curricula to Cultural Awareness: Constructing Open-Ended Culture-Specific Question Answering Dataset

arXiv:2601.04632v2 Announce Type: replace Abstract: Large language models (LLMs) achieve strong performance on many tasks, but their progress remains uneven across languages and cultures, often reflec

From Preferences to Principles: Rubric-Based Alignment for Grounded Knowledge Answers

Designing effective reward signals for open-domain question answering is challenging because high-quality responses must simultaneously satisfy multiple aspects of answer quality that are difficult to

safetyapple-ml-research

GaussianDream++: Efficient 3D Gaussian World Modeling for Robotic Manipulation

arXiv:2608.25659v1 Announce Type: new Abstract: Vision-Language-Action (VLA) policies have advanced language-conditioned robotic manipulation, yet action-imitation objectives provide only weak supervi

Generative Action-Chunk Sampling for Adaptive Stiffness Control in Physical Human-Robot Collaboration

arXiv:2608.25284v1 Announce Type: cross Abstract: Physical human-robot collaboration requires a robot to provide assistance when human intention is clear while remaining compliant when several future

GGSS: Geodesic-Gated Spherical Steering for Inference-Time Debiasing of Generative Vision-Language Models

arXiv:2608.25375v1 Announce Type: cross Abstract: Generative vision-language models (VLMs) are increasingly used in human-centered settings, yet they can produce demographically biased outputs even wh

Imitation Learning for Connection-Tableau Construction

arXiv:2608.26009v1 Announce Type: cross Abstract: An automated theorem prover builds a proof step by step, choosing at each point what to add and what to remove. We cast this construction as a policy

Improving Cross-Site Whole-Heart Segmentation

arXiv:2608.25109v1 Announce Type: cross Abstract: Whole-heart segmentation from CT and MRI is essential for quantitative cardiac image analysis, but remains challenging under multi-center and multi-mo

Individual Fairness in Hierarchical Clustering

arXiv:2608.25586v1 Announce Type: new Abstract: Hierarchical clustering produces ultrametric representations that impose strong global geometric constraints and may distort local similarities in ways

Large Language Model Few-Shot Prompting with Dilemma Training Outperforms Human Surrogates in Predicting Patient Preferences

arXiv:2608.25771v1 Announce Type: cross Abstract: In serious illness, human surrogates often struggle to accurately predict patient preferences (68% accuracy), causing decision conflict. Personalized

Learning Late, Guiding Early: Timestep-Decoupled Semantic Guidance for Fair Face Generation

arXiv:2608.25862v1 Announce Type: new Abstract: Demographic imbalance in synthetic face generation can propagate to downstream face recognition systems, making fairness an important consideration when

Learning Mixtures of Plackett-Luce Models for Multi-Objective Alignment

arXiv:2608.25200v1 Announce Type: cross Abstract: We consider the problem of learning a mixture of k Plackett-Luce models given multi-way ranking responses from annotators that may represent heterogen

Learning to Accelerate Vision-Language-Action Models through Adaptive Visual Token Caching

arXiv:2602.00686v2 Announce Type: replace Abstract: Vision-Language-Action (VLA) models have demonstrated remarkable generalization capabilities in robotic manipulation tasks, yet their substantial co

Learning What to Share and What to Personalize: Hierarchical Strategy Co-Evolution for Agent Memory

arXiv:2608.25329v1 Announce Type: cross Abstract: Memory-augmented agents maintain compact user profiles throughout extended conversations, enabling personalized and consistent responses without the n

Lightweight Machine Learning-Driven Monocular Sidewalk Path Extraction for Embedded Micromobility Navigation

arXiv:2608.25178v1 Announce Type: new Abstract: Sidewalk-scale path extraction demands perception and planning that run reliably on compact, low-power hardware in cluttered, map-sparse environments. W

Most companies manage their org chart, their product line, their brand guidelines, their operations manual. Each as a separate document. Eac…

Most companies manage their org chart, their product line, their brand guidelines, their operations manual. Each as a separate document. Each updated on its own cadence. Each a snapshot of a different

safetyyohei-nakajima--x

Neither Precision Nor Architecture Alone: Controlled Tests of Failure Remedies for Physics-Informed Neural Networks

arXiv:2608.25327v1 Announce Type: new Abstract: Physics-Informed Neural Networks (PINNs) frequently fail on stiff or advection-dominated PDEs, and two recent accounts offer competing remedies: switchi

Neural-Bayesian Structure Learning for Discrete Choice Modeling

arXiv:2608.25258v1 Announce Type: new Abstract: Conventional discrete choice and machine learning models are estimated primarily from observational data and typically treat explanatory covariates as p

One Symptom, Three Levers: A Critical Review of On-Policy Self-Distillation

arXiv:2608.25936v1 Announce Type: cross Abstract: On-policy distillation trains a language model on its own generations while a teacher scores them token by token. It combines the dense supervision of

PAGS: Autofocusing Photoacoustic Tomography via Speed-of-Sound-Adaptive Gaussian Splatting

arXiv:2608.25472v1 Announce Type: new Abstract: Photoacoustic computed tomography (PACT) combines optical absorption contrast with acoustic detection for high-resolution deep-tissue imaging. A persist

PANDA - Prototype-Anchored Alignment for Partially Unpaired Multimodal Learning, with Applications to Alzheimers MRI and TCGA Pathology

arXiv:2608.25970v1 Announce Type: new Abstract: Multimodal medical prediction often faces incomplete pairing: auxiliary modalities with complementary signal are available for only a subset of subjects

Phase-Consistent Magnetic Spectral Learning for Multi-View Clustering

arXiv:2602.18728v2 Announce Type: replace-cross Abstract: Unsupervised multi-view clustering (MVC) aims to partition data into meaningful groups by leveraging complementary information from multiple v

PreResQ-R1: Response-Preference Disentangled Ranking-and-Scoring Reinforcement Optimization for Robust Visual Quality Assessment

arXiv:2511.05393v2 Announce Type: replace Abstract: Visual Quality Assessment (QA) seeks to predict human perceptual judgments of visual fidelity. While recent multimodal large language models (MLLMs)

Provable Privacy Attacks on Trained Shallow Neural Networks

arXiv:2410.07632v3 Announce Type: replace Abstract: We study what provable privacy attacks can be shown for trained 2-layer ReLU neural networks, focusing on two types of attacks: membership inference

ReliableRAG: Combating Misinformation in Retrieval-Augmented Generation via Reliability-Guided Reasoning Chains

arXiv:2608.25487v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) has emerged as a powerful architecture for Question Answering (QA) by integrating external information into Large L

Retrieved But Not Reliable: A Survey on Attacks, and Defenses in Retrieval-Augmented Generation

arXiv:2608.24977v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) enhances large language models by grounding outputs in external knowledge, improving factuality and reducing hall

SEG-SAM: Semantic-Guided SAM for Unified Medical Image Segmentation

arXiv:2412.12660v2 Announce Type: replace Abstract: Recently, developing unified medical image segmentation models gains increasing attention, especially with the advent of the Segment Anything Model

Semantic Variability of Replies Across LLMs: Implications for Designing Conversation-Based Assessment

arXiv:2608.24920v1 Announce Type: new Abstract: This study examines whether LLM-generated replies remain semantically consistent when the underlying LLM changes. Using messages from real collaborative

Several articles, podcasts etc now about the Open AI Hugging Face incident, from reputable outlets, framing this as 'AI going rouge' & debat…

Several articles, podcasts etc now about the Open AI Hugging Face incident, from reputable outlets, framing this as 'AI going rouge' & debating what AI did or didn't do. When can we shift to focusing

safetygary-marcus--x

Short Horizons and Sparse Concepts: a Mathematical View of the Readout in the J-lens

arXiv:2608.25347v1 Announce Type: new Abstract: The Jacobian lens (J-lens) has been proposed as a way to read verbalizable representations from language models. However, its principle and meaning lack

ShuttleArena: Interpretable Self-Play in Physics-Based Badminton

arXiv:2608.25246v1 Announce Type: new Abstract: Badminton is a compact but challenging domain for game AI: a player must choose a physically feasible shuttle trajectory, anticipate the opponent's inte

Socialized Detector Learning: Trajectory-Guided and Reciprocal Distillation for Heterogeneous Object Detectors

arXiv:2608.25836v1 Announce Type: new Abstract: Object detection knowledge is fragmented across independently trained, heterogeneous detectors with complementary category supports. In socialized learn

SonicNudge: Controlled Displacement of Hovering UAVs via Estimator-Controller Coupling

arXiv:2608.25319v1 Announce Type: cross Abstract: UAV displacement attacks have traditionally relied on spoofing sensors that directly report position or translational motion, such as GNSS and optical

TDFNet: Tri-projection Deformable Fusion Network for Panoramic Salient Object Detection

arXiv:2608.25808v1 Announce Type: new Abstract: Recent years have witnessed the growing potential of panoramic salient object detection in robotic vision, virtual reality, and related applications. Ho

The coolest part of Microduck might not be the robot. It's how it learns to move. This repo shows the full reinforcement learning pipeline b…

The coolest part of Microduck might not be the robot. It's how it learns to move. This repo shows the full reinforcement learning pipeline behind Microduck. The robot learns in simulation using MuJoCo

safetyclem-delangue--x

Thermodynamic cost of inference and learning in physical neural networks

arXiv:2503.09980v4 Announce Type: replace-cross Abstract: How much of the energy consumed by artificial neural networks is set by physics rather than by implementation? For irreversible digital hardwa

Think-Probe-Respond: Improving Large Language Models as Judges of Research Idea Novelty

arXiv:2608.25660v1 Announce Type: new Abstract: Automated novelty judgment can accelerate scientific discovery by enabling efficient evaluation, refinement, and comparison of research ideas. While lar

Token-Oriented Semantic Communication with Pretrained Vision Transformers

arXiv:2608.25410v1 Announce Type: cross Abstract: Token communications realize the semantic communication principle at the granularity of transformer tokens, providing a promising direction for client

Towards Purified Multi-Label Test-Time Adaptation of Vision-Language Models

arXiv:2608.25653v1 Announce Type: new Abstract: Test-time adaptation (TTA) has been widely explored in single-label recognition, effectively mitigating distribution shifts, especially when combined wi

Training Alignment Auditors via Reinforcement Learning

arXiv:2608.25460v1 Announce Type: cross Abstract: Alignment auditing of frontier models increasingly relies on LLM auditors to surface undesirable behaviors at scale, but current automated auditors ca

Training-Free Interaction-Aligned Visual Token Pruning for Efficient Embodied Manipulation

arXiv:2603.22991v2 Announce Type: replace Abstract: Efficient visual representation is a central image-processing challenge in embodied manipulation, where policies repeatedly process dense visual-tok

VBVR-Pro: A Scalable and Verifiable Suite for Native Visual Reasoning

arXiv:2608.26105v1 Announce Type: new Abstract: Native visual reasoning treats visual generation as the medium of reasoning itself: visual states (i.e. images and videos) are not merely inputs to be u

VISTA: Visually Inferred Spatial ConTact Attention for Contact-Rich Manipulation

arXiv:2608.25872v1 Announce Type: new Abstract: Contact-rich manipulation requires precise interaction feedback. While vision-centric imitation learning is prevalent, external visual observations prov

WAVE: Reversing the Guidance Hierarchy for Coarse-to-Fine Guided Depth Super-Resolution

arXiv:2608.25302v1 Announce Type: new Abstract: Guided depth super-resolution (GDSR) typically extracts RGB guidance features through convolutional hierarchies, inheriting their fine-to-coarse bias. T

What Do Medical Vision-Language Models Learn in Radiology? Transfer, Alignment, and Source-Proxy Leakage Under Distribution Shift

arXiv:2608.25251v1 Announce Type: new Abstract: Medical vision-language models (VLMs) can appear reliable in-domain while failing when acquisition domain, paired supervision, or evaluation protocol ch