127,664Total entries

60Added by human

127,604Found by agent

13Categories

Knowledge catalogue

Search: “safety”

22,398 results

Safety

i-WiViG: Interpretable Window Vision GNN

DGX agent

arXiv:2503.08321v2 Announce Type: replace Abstract: Vision graph neural networks have emerged as a popular approach for modeling the global and spatial context for image recognition. However, a signif

safetyarxiv-cs-cv

23 Apr 2026

Safety

Search: “safety”

i-WiViG: Interpretable Window Vision GNN

I’m in Madrid this week for the first in-person meeting of the United Nations Independent International Scientific Panel on AI. I'm thrilled…

Kalman Filter Enhanced GRPO for Reinforcement Learning-Based Language Model Reasoning

Language-Coupled Reinforcement Learning for Multilingual Retrieval-Augmented Generation

Learning to count small and clustered objects with application to bacterial colonies

Lever: Inference-Time Policy Reuse under Support Constraints

LLM Agents Predict Social Media Reactions but Do Not Outperform Text Classifiers: Benchmarking Simulation Accuracy Using 120K+ Personas of 1511 Humans

MATT-Diff: Multimodal Active Target Tracking by Diffusion Policy

Membership Inference for Contrastive Pre-training Models with Text-only PII Queries

Memorization, Emergence, and Explaining Reversal Failures: A Controlled Study of Relational Semantics in LLMs

MOA: Multi-Objective Alignment for Role-Playing Agents

Multi-Armed Bandits With Machine Learning-Generated Surrogate Rewards

Multi-Objective Reinforcement Learning for Generating Covalent Inhibitor Candidates

Near-Future Policy Optimization

Occupancy Reward Shaping: Improving Credit Assignment for Offline Goal-Conditioned Reinforcement Learning

ORPHEAS: A Cross-Lingual Greek-English Embedding Model for Retrieval-Augmented Generation

Participatory provenance as representational auditing for AI-mediated public consultation

Peer-Preservation in Frontier Models

ProMMSearchAgent: A Generalizable Multimodal Search Agent Trained with Process-Oriented Rewards

Recency Biased Causal Attention for Time-series Forecasting

Relative Principals, Pluralistic Alignment, and the Structural Value Alignment Problem

Representational Alignment Across Model Layers and Brain Regions with Multi-Level Optimal Transport

Rethinking Reinforcement Fine-Tuning in LVLM: Convergence, Reward Decomposition, and Generalization

Rodrigues Network for Learning Robot Actions

Sampling-Aware Quantization for Diffusion Models

SignDATA: Data Pipeline for Sign Language Translation

Storm Surge Modeling, Bias Correction, Graph Neural Networks, Graph Convolution Networks

Temporal Difference Calibration in Sequential Tasks: Application to Vision-Language-Action Models

The Existential Theory of Research: Why Discovery Is Hard

The Imperfective Paradox in Large Language Models

The Tool-Overuse Illusion: Why Does LLM Prefer External Tools over Internal Knowledge?

Throat and acoustic paired speech dataset for deep learning-based speech enhancement

Understanding Overparametrization in Survival Models through Interpolation

UVIO: An UWB-Aided Visual-Inertial Odometry Framework with Bias-Compensated Anchors Initialization

V-tableR1: Process-Supervised Multimodal Table Reasoning with Critic-Guided Policy Optimization

Visual-Tactile Peg-in-Hole Assembly Learning from Peg-out-of-Hole Disassembly

What Makes a Good AI Review? Concern-Level Diagnostics for AI Peer Review

Where and What: Reasoning Dynamic and Implicit Preferences in Situated Conversational Recommendation

Which Reasoning Trajectories Teach Students to Reason Better? A Simple Metric of Informative Alignment

Whose Story Gets Told? Positionality and Bias in LLM Summaries of Life Narratives

Working Memory Constraints Scaffold Learning in Transformers under Data Scarcity

A UK tribunal rules Microsoft must face a lawsuit alleging it overcharged UK businesses to run Windows Server on cloud services from Amazon, Google, and Alibaba (Sam Tobin/Reuters)

Adaptive Prompt Elicitation for Text-to-Image Generation

AeroBridge-TTA: Test-Time Adaptive Language-Conditioned Control for UAVs

AI failure could trigger the next financial crisis, warns Elizabeth Warren

AlignedCut: Visual Concepts Discovery on Brain-Guided Universal Feature Space

Allo{SR}^2: Rectifying One-Step Super-Resolution to Stay Real via Allomorphic Generative Flows

Anthropic’s own internal security blows.