Search: “safety”

The New York Times and several news outlets are accusing OpenAI of hiding its ability to search its AI training data and output logs for mor…

The New York Times and several news outlets are accusing OpenAI of hiding its ability to search its AI training data and output logs for more than two years during an ongoing copyright lawsuit: • The

safetygary-marcus--x

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

Thinking Seeds: Leveraging Historical Diversity for Position-Aware RL in LLMs

DGX agent

arXiv:2601.21476v2 Announce Type: replace Abstract: On-policy reinforcement learning (RL) for language model post-training suffers from a fundamental tension: as training progresses, policy entropy co

safetyarxiv-cs-cl

9 Jul 2026

This is a very important observation. AI is not adopting the language and shaping it to fit its known ideas. It has no known ideas. It is re…

This is a very important observation. AI is not adopting the language and shaping it to fit its known ideas. It has no known ideas. It is repeating patterns found in the language. super interesting -

safetygary-marcus--x

TIR-Agent: Training an Explorative and Efficient Agent for Image Restoration

arXiv:2603.27742v2 Announce Type: replace Abstract: Vision-language agents that orchestrate specialized tools for image restoration (IR) have emerged as a promising method, yet most existing framework

Towards a Pseudo-Labeling Workflow for Celltype-Classification from Explanted Brain Slice Recordings

arXiv:2607.06569v1 Announce Type: cross Abstract: This paper proposes an unsupervised workflow to pseudo-label extracellular spikes from human brain slice MEA recordings into two putative cell types:

Trajectory Inference of Human Aging from Cross-Sectional DNA Methylation Data

arXiv:2607.06583v1 Announce Type: cross Abstract: DNA methylation (DNAm) serves as one of the most robust molecular biomarkers of biological aging. While conventional epigenetic clocks accurately pred

Unmasking On-Policy Distillation: Where It Helps, Where It Hurts, and Why

On-policy distillation offers dense, per-token supervision for training reasoning models; however, it remains unclear under which conditions this signal is beneficial and under which it is detrimental

safetyapple-ml-research

UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma

arXiv:2607.06987v1 Announce Type: new Abstract: Reinforcement learning (RL) has become the standard paradigm for enhancing the complex reasoning capabilities of large language models (LLMs). To achiev

VFM-Loc: Training-Free Cross-View Geo-Localization via Aligning Discriminative Visual Hierarchies

arXiv:2603.13855v2 Announce Type: replace Abstract: Cross-View Geo-Localization (CVGL) in remote sensing aims to locate a drone-view query by matching it to geo-tagged satellite images. Although super

Vision Foundation Models in Radiology: A Scoping Review of Data, Methodology, Evaluation and Clinical Translation

arXiv:2607.07219v1 Announce Type: cross Abstract: Vision foundation models (VFMs) are increasingly being developed for radiological imaging, yet their definition, development and evaluation remain het

When Agents Remember Too Much: Memory Poisoning Attacks on Large Language Model Agents

arXiv:2607.06595v1 Announce Type: cross Abstract: Personal AI agents powered by large language models can reason and act using available tools to access emails, manage calendars, and push code to remo

WHERE to Generate Matters: Budget-Aware Synthetic Augmentation for Label Skewed Federated Learning

arXiv:2607.06616v1 Announce Type: cross Abstract: Label skew in federated learning (FL) causes client drift and degrades global accuracy. Synthetic data augmentation can reduce this imbalance; however

Zoom In Disparities in Healthcare LLM Q&A

arXiv:2510.17476v2 Announce Type: replace Abstract: Equitable access to reliable health information is vital when integrating AI into healthcare. Yet, information quality varies across languages, rais

safetyarxiv-cs-cl

A Coin Flip Per Token: Bernoulli Sparse Steering of Large Language Models

arXiv:2607.05615v1 Announce Type: new Abstract: Activation steering via sparse autoencoders (SAEs) enables behavioral control of large language models without task-specific fine-tuning, but standard m

A Function-Space Dichotomy for Compositional Learning: Exponential Sub-Optimality of the Neural Tangent Kernel

arXiv:2607.06382v1 Announce Type: cross Abstract: A persistent empirical observation is that trained neural networks outperform their neural tangent kernel (NTK) limit on tasks with compositional stru

A toy framework for single and multi-agent human-AI curiosity ecosystems

arXiv:2607.06214v1 Announce Type: new Abstract: This paper offers a toy framework for considering curiosity as an ecosystem. First, it suggests that a single agent's inquiry policy (how, when, and why

Model Releases

A VLM-Enhanced Framework for Comprehensive Traffic Sign Condition Assessment Integrating Daytime Visual Performance and Nighttime Retroreflectivity Evaluation

model-releasesarxiv-cs-cv

arXiv:2607.06478v1 Announce Type: new Abstract: Traffic signs are crucial components of road safety, serving as visual tools under all lighting conditions. The Manual on Uniform Traffic Control Device

ARMS: Anchor-Relational Motion Streaming for Seamless Solo-Social Motion Transitions

arXiv:2607.05733v1 Announce Type: new Abstract: Generating temporally continuous and socially coherent human motion from text remains a fundamental challenge, particularly in realistic streams where p

Automated Recommendation of Programming Learning Content Using Pattern-based Knowledge Components

arXiv:2607.05409v1 Announce Type: cross Abstract: Introductory programming instruction relies on hands-on practice and short learning activities to support mastery of foundational concepts. Although m

Model Releases

Beyond Refusal: A Same-Lineage Study of Aligned and Abliterated LLMs for Vulnerability Analysis

model-releasesarxiv-cs-ai

arXiv:2607.05842v1 Announce Type: cross Abstract: Large language model (LLM)-assisted software security operates at a difficult boundary: the vulnerability-analysis terminology needed for legitimate c

Bridging Diffusion Pruning and Step Distillation with Teacher-Aligned Repair

arXiv:2607.06335v1 Announce Type: new Abstract: Diffusion models generate high-quality images, but their inference cost comes from two sources: large denoising networks and repeated denoising steps. E

Bridging Physical Reasoning and Task Generalization via Visual Action Outcome Reasoning Alignment

arXiv:2607.06522v1 Announce Type: new Abstract: Vision-language models (VLMs) struggle to generalize in interactive physical reasoning, particularly under unseen tasks and environments. Two key failur

Choose What to Observe: Task-Aware Semantic-Geometric Representations for Visuomotor Policy

arXiv:2603.07875v2 Announce Type: replace Abstract: Visuomotor policies learned from demonstrations often overfit to nuisance visual factors in raw RGB observations, resulting in brittle behavior unde

safetyarxiv-cs-ro

Controlling Tool Use with Heading-Specific Activation Steering

arXiv:2607.05790v1 Announce Type: new Abstract: Tool-augmented large language models extend their capabilities beyond parametric knowledge through external tools, but tend to invoke them unnecessarily

Cross-Contextual Vision-Language Adaptation with LoRA for Personalized Severe Adverse Event Detection in Clinical Wound Monitoring

arXiv:2607.05625v1 Announce Type: new Abstract: Wound monitoring is a critical yet underserved clinical challenge, where timely identification of severe adverse events (SAEs) such as infection, tissue

Deep Reinforcement Learning for Dynamic Origin-Destination Matrix Estimation in Microscopic Traffic Simulations Considering Credit Assignment

arXiv:2511.06229v3 Announce Type: replace Abstract: This paper focuses on dynamic origin-destination matrix estimation (DODE), a crucial calibration process necessary for the effective application of

DeSeG: Decoupling Semantic Intent and Geometric Constraints for Physically Plausible Human-Scene Interaction

arXiv:2607.05787v1 Announce Type: new Abstract: Synthesizing physically plausible human-scene interactions (HSI) remains a critical challenge in computer vision and the development of human avatars. A

Domain-Adaptive Climate Downscaling Under Temporal Distribution Shift

arXiv:2607.05645v1 Announce Type: new Abstract: Deep-learning-based climate downscaling aims to learn relationships from historical low-resolution (LR) and high-resolution (HR) climate data to generat

DreamPartGen: Semantically Grounded Part-Level 3D Generation via Collaborative Latent Denoising

arXiv:2603.19216v2 Announce Type: replace-cross Abstract: Understanding and generating 3D objects as compositions of meaningful parts is fundamental to human perception and reasoning. However, most te

DynaKRAG: A Unified Framework for Learnable Evidence Control in Multi-Hop Retrieval-Augmented Generation

arXiv:2607.06507v1 Announce Type: new Abstract: Multi-hop retrieval-augmented generation (RAG) acquires evidence sequentially, with each new document potentially revealing missing facts, bridge entiti

safetyarxiv-cs-cl

ELSA3D: Elastic Semantic Anchoring for Unified 3D Understanding and Generation

arXiv:2607.06565v1 Announce Type: cross Abstract: Unified 3D foundation models aspire to generate 3D assets and reason about them in language within a single backbone, but their text-3D interaction re

Embodied Human-Robot Interaction via Acoustics: A MARL Approach with AcoustoBots for Spatial Data Physicalization

arXiv:2607.06563v1 Announce Type: new Abstract: Traditional data physicalization is often static and disconnected from real environments, limiting its ability to convey embodied spatial dynamics and e

safetyarxiv-cs-ro

FADRA: Frequency-Aware Diffusion with Residual Adaptation for Video Face Restoration

arXiv:2607.06389v1 Announce Type: new Abstract: Video face restoration (VFR) aims to recover high-quality and temporally consistent facial details from severely degraded video sequences; however, exis

FreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference

arXiv:2607.06519v1 Announce Type: new Abstract: Long-context LLM inference is increasingly limited by the memory and bandwidth cost of KV caches, yet aggressive compression can remove the layer-specif

From Application-Layer Simulation to Native Meta-Architecture: Structural Tension as an Endogenous Driver for Heterogeneous AI Evolution

arXiv:2607.06269v1 Announce Type: new Abstract: Current large language models (LLMs) are fundamentally stateless: their behavior is fully determined by input at inference time, and any higher-order co

From Blueprint to Reality: Modeling and Applying Putnam's Social Capital Theory with LLM-based Multi-agent Simulations

arXiv:2607.06080v1 Announce Type: cross Abstract: Putnam's Social Capital Theory is a foundational framework for collective action and community prosperity. However, traditional empirical methods face

From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space

arXiv:2607.05794v1 Announce Type: new Abstract: Long-term user memory is essential for personalized conversational agents, yet many memory systems still expose memory through passive retrieval interfa

From Pixels to Portraits: A Comprehensive Survey of Talking Head Generation Techniques and Applications

arXiv:2308.16041v2 Announce Type: replace Abstract: Talking head generation has progressed rapidly from landmark- and GAN-based facial animation to diffusion models, neural rendering, 3D-aware avatars

GaussFusion: Towards Multimodal 3D Gaussian Pretraining

arXiv:2607.05906v1 Announce Type: new Abstract: 3D Gaussian Splatting provides an explicit representation that jointly models geometry and appearance, serving as a scalable foundation for 3D represent

GraspIT: A Dataset Bridging the Sim-to-Real gap and back for Validated Grasping SE(3) Pose Generation

arXiv:2607.05869v1 Announce Type: cross Abstract: Robust robotic grasping of novel objects requires datasets that simultaneously provide photorealistic RGB-D observations, physically validated grasp q

Heckman-Corrected Epistemic Uncertainty: Selection on Unobservables Defeats Importance Weighting

arXiv:2607.05806v1 Announce Type: new Abstract: Training data for machine learning is routinely collected by a selection process the model never sees: loans are observed only when granted, outcomes on

Information Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn LLM Agents

arXiv:2607.06223v1 Announce Type: new Abstract: Reinforcement learning has become a promising paradigm for improving large language model (LLM) agents on long-horizon search tasks, where the agent mus

Intercepting an Agile Target with Net-Carrying Drones using Competitive Multi-Agent Reinforcement Learning

arXiv:2607.05939v1 Announce Type: new Abstract: This article presents a solution to intercept an agile drone by a team of agile drone carrying catching nets. We formulate the problem as a competitive

safetyarxiv-cs-ro

Is Domain Adaptation Always Helpful? A Frozen-Backbone Study of Cross-Domain Sentiment Transfer

arXiv:2607.05937v1 Announce Type: new Abstract: Sentiment analysis with frozen pre-trained language model (PLM) backbones has become a common paradigm, yet the practical benefit of explicit domain ada

safetyarxiv-cs-cl

IUU+DB: Tracking Illegal, Unreported, and Unregulated Fishing, Seafood Fraud, and Labor Abuse through LLM-driven Information Extraction

arXiv:2606.18181v2 Announce Type: replace-cross Abstract: Illegal, unreported, and unregulated fishing (IUU) traditionally refers to fishing activities that violate applicable laws or occur in areas t

K-ABENA: K-Adaptive Backpropagation with Error-based N-exclusion Algorithm : (Compensated Loss-Based Sample Exclusion with Unbiased Gradient Estimation)

arXiv:2607.05903v1 Announce Type: cross Abstract: We present K-ABENA (K-Adaptive Backpropagation with Error-based N-exclusion Algorithm), a selective gradient computation framework that reduces per-it

KAT-Coder-V2.5 Technical Report

arXiv:2607.05471v1 Announce Type: cross Abstract: We present KAT-Coder-V2.5, a coding-focused agentic model trained to act autonomously inside real, executable repositories rather than as a single-tur

KernelEvolve: Scaling Agentic Kernel Coding for Heterogeneous AI Accelerators at Meta

arXiv:2512.23236v4 Announce Type: replace-cross Abstract: Making deep learning recommendation model (DLRM) training and inference fast and efficient is important. However, this presents three key syst