Search: “safety”

Learning from Medical Entity Trees: An Entity-Centric Medical Data Engineering Framework for MLLMs

arXiv:2604.25296v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) have shown transformative potential in medical applications, yet their performance is hindered by conventional

29 Apr 2026

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

Sustained Gradient Alignment Mediates Subliminal Learning in a Multi-Step Setting: Evidence from MNIST Auxiliary Logit Distillation Experiment

DGX agent

arXiv:2604.25779v1 Announce Type: new Abstract: In the MNIST auxiliary logit distillation experiment, a student can acquire an unintended teacher trait despite distilling only on no-class logits throu

safetyarxiv-cs-lg

29 Apr 2026

Vocabulary Dropout for Curriculum Diversity in LLM Co-Evolution

arXiv:2604.03472v2 Announce Type: replace Abstract: Co-evolutionary self-play, where one language model generates problems and another solves them, promises autonomous curriculum learning without huma

29 Apr 2026

Adaptive Multi-Subspace Representation Steering for Attribute Alignment in Large Language Models

arXiv:2508.10599v4 Announce Type: replace Abstract: Activation steering offers a promising approach to controlling the behavior of Large Language Models by directly manipulating their internal activat

AMAVA: Adaptive Motion-Aware Video-to-Audio Framework for Visually-Impaired Assistance

arXiv:2604.23909v1 Announce Type: new Abstract: Navigational aids for blind and low vision individuals struggle conveying dynamic real-world environments, leading to cognitive overload from continuous

safetyarxiv-cs-cv

An empirical evaluation of the risks of AI model updates using clinical data: stability, arbitrariness, and fairness

arXiv:2604.23954v1 Announce Type: new Abstract: Artificial Intelligence and Machine Learning (AI/ML) models used in clinical settings are increasingly deployed to support clinical decision-making. How

Analytica: Soft Propositional Reasoning for Robust and Scalable LLM-Driven Analysis

arXiv:2604.23072v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly tasked with complex real-world analysis (e.g., in financial forecasting, scientific discovery), yet t

ArgRE: Formal Argumentation for Conflict Resolution in Multi-Agent Requirements Negotiation

arXiv:2604.23124v1 Announce Type: cross Abstract: As software systems grow in complexity, they must satisfy an increasing number of competing quality attributes, making it essential to balance them in

Autocorrelation Reintroduces Spectral Bias in KANs for Time Series Forecasting

arXiv:2604.23518v1 Announce Type: cross Abstract: Existing theory suggests that Kolmogorov-Arnold Networks (KANs) can overcome the spectral bias commonly observed in neural networks under the assumpti

Model Releases

Beyond Context: Large Language Models' Failure to Grasp Users' Intent

model-releasesarxiv-cs-ai

arXiv:2512.21110v3 Announce Type: replace Abstract: Current Large Language Models (LLMs) safety approaches focus on explicitly harmful content while overlooking a critical vulnerability: the inability

Context-Aware Hospitalization Forecasting Evaluations for Decision Support using LLMs

arXiv:2604.23949v1 Announce Type: new Abstract: Medical and public health experts must make real-time resource decisions, such as expanding hospital bed capacity, based on projected hospitalization tr

Designing escalation criteria for international AI incident response: criteria, triggers, and thresholds

arXiv:2604.23183v1 Announce Type: cross Abstract: AI incident reporting requirements are emerging in regulation and policy, yet no operational criteria exist for determining when a detected AI inciden

Designing Instance-Level Sampling Schedules via REINFORCE with James-Stein Shrinkage

arXiv:2511.22177v2 Announce Type: replace-cross Abstract: Most post-training methods for text-to-image samplers focus on model weights: either fine-tuning the backbone for alignment or distilling it f

safetyarxiv-cs-cv

From Stateless Queries to Autonomous Actions: A Layered Security Framework for Agentic AI Systems

arXiv:2604.23338v1 Announce Type: cross Abstract: Agentic AI systems face security challenges that stateless large language models do not. They plan across extended horizons, maintain persistent memor

Governing What You Cannot Observe: Adaptive Runtime Governance for Autonomous AI Agents

arXiv:2604.24686v1 Announce Type: new Abstract: Autonomous AI agents can remain fully authorized and still become unsafe as behavior drifts, adversaries adapt, and decision patterns shift without any

LLM-Auction: Generative Auction towards LLM-Native Advertising

arXiv:2512.10551v2 Announce Type: replace-cross Abstract: The commercialization of LLM applications is the next frontier in online advertising, with LLM-native advertising emerging as a promising para

Polychromic Objectives for Reinforcement Learning

arXiv:2509.25424v5 Announce Type: replace-cross Abstract: Reinforcement learning fine-tuning (RLFT) is a dominant paradigm for improving pretrained policies for downstream tasks. These pretrained poli

Position: Logical Soundness is not a Reliable Criterion for Neurosymbolic Fact-Checking with LLMs

arXiv:2604.04177v2 Announce Type: replace Abstract: As large language models (LLMs) are increasing integrated into fact-checking pipelines, formal logic is often proposed as a rigorous means by which

ProEval: Proactive Failure Discovery and Efficient Performance Estimation for Generative AI Evaluation

arXiv:2604.23099v1 Announce Type: cross Abstract: Evaluating generative AI models is increasingly resource-intensive due to slow inference, expensive raters, and a rapidly growing landscape of models

Protecting the Trace: A Principled Black-Box Approach Against Distillation Attacks

arXiv:2604.23238v1 Announce Type: cross Abstract: Frontier models push the boundaries of what is learnable at extreme computational costs, yet distillation via sampling reasoning traces exposes closed

Safe Navigation in Unknown and Cluttered Environments via Direction-Aware Convex Free-Region Generation

arXiv:2604.23648v1 Announce Type: new Abstract: Convex free regions provide a structured and optimization-friendly representation of collision-free space for robot navigation in unknown and cluttered

safetyarxiv-cs-ro

Security Considerations for Multi-agent Systems

arXiv:2603.09002v2 Announce Type: replace-cross Abstract: Multi-agent artificial intelligence systems or MAS are systems of autonomous agents that exercise delegated tool authority, share persistent m

Sliding Mode Control for Safe Trajectory Tracking with Moving Obstacles Avoidance: Experimental Validation on Planar Robots

arXiv:2604.24518v1 Announce Type: cross Abstract: This paper presents a unified control framework for robust trajectory tracking and moving obstacle avoidance applicable to a broad class of mobile rob

safetyarxiv-cs-ro

Synthetic Eggs in Many Baskets: The Impact of Synthetic Data Diversity on LLM Fine-Tuning

arXiv:2511.01490v2 Announce Type: replace Abstract: As synthetic data becomes widely used in language model development, understanding its impact on model behavior is crucial. This paper investigates

Training a General Purpose Automated Red Teaming Model

arXiv:2604.23067v1 Announce Type: cross Abstract: Automated methods for red teaming LLMs are an important tool to identify LLM vulnerabilities that may not be covered in static benchmarks, allowing fo

Value Alignment Tax: Measuring Value Trade-offs in LLM Alignment

arXiv:2602.12134v2 Announce Type: replace Abstract: Existing work on value alignment typically characterizes value relations statically, ignoring how alignment interventions, such as prompting, fine-t

Adaptive vs. Static Robot-to-Human Handover: A Study on Orientation and Approach Direction

arXiv:2604.22378v1 Announce Type: new Abstract: Robot-to-human handovers often rely on static, open-loop strategies (or, at best, approaches that adapt only the position), which generally do not consi

safetyarxiv-cs-ro

AgentBound: Securing Execution Boundaries of AI Agents

arXiv:2510.21236v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) have evolved into AI agents that interact with external tools and environments to perform complex tasks. The Mode

Controllable Spoken Dialogue Generation: An LLM-Driven Grading System for K-12 Non-Native English Learners

arXiv:2604.22542v1 Announce Type: cross Abstract: Large language models (LLMs) often fail to meet the pedagogical needs of K-12 English learners in non-native contexts due to a proficiency mismatch. T

Don't try to build a self-improving AI agent without evals. You are just wasting time and compute. An agent can't improve from traces it can…

Don't try to build a self-improving AI agent without evals. You are just wasting time and compute. An agent can't improve from traces it can't evaluate. This is why it's exciting to see @FutureAGI_ go

safetydair-ai--x

Emergent Strategic Reasoning Risks in AI: A Taxonomy-Driven Evaluation Framework

arXiv:2604.22119v1 Announce Type: new Abstract: As reasoning capacity and deployment scope grow in tandem, large language models (LLMs) gain the capacity to engage in behaviors that serve their own ob

Sam Altman cannot be trusted. • The OpenAI board fired him because he was not always honest with them. They said he should not control power…

Sam Altman cannot be trusted. • The OpenAI board fired him because he was not always honest with them. They said he should not control powerful AI. • A major report talked to over 100 people and saw s

safetyelon-musk--x

Bounding the Black Box: A Statistical Certification Framework for AI Risk Regulation

arXiv:2604.21854v1 Announce Type: new Abstract: Artificial intelligence now decides who receives a loan, who is flagged for criminal investigation, and whether an autonomous vehicle brakes in time. Go

Enabling and Inhibitory Pathways of University Students' Willingness to Disclose AI Use: A Cognition-Affect-Conation Perspective

arXiv:2604.21733v1 Announce Type: new Abstract: The increasing integration of artificial intelligence (AI) in higher education has raised important questions regarding students' transparency in report

HARBOR: Automated Harness Optimization

arXiv:2604.20938v1 Announce Type: cross Abstract: Long-horizon language-model agents are dominated, in lines of code and in operational complexity, not by their underlying model but by the harness tha

Lawmakers and lobbyists say the Trump administration has lobbied against legislation that would regulate AI in at least six Republican-led states (Amrith Ramkumar/Wall Street Journal)

Amrith Ramkumar / Wall Street Journal: Lawmakers and lobbyists say the Trump administration has lobbied against legislation that would regulate AI in at least six Republican-led states — ‘I am disappo

safetytechmeme

Measuring and Exploiting Contextual Bias in LLM-Assisted Security Code Review

arXiv:2603.18740v2 Announce Type: replace-cross Abstract: Automated Code Review (ACR) systems integrating Large Language Models (LLMs) are increasingly adopted in software development workflows, rangi

Mind the Prompt: Self-adaptive Generation of Task Plan Explanations via LLMs

arXiv:2604.21092v1 Announce Type: new Abstract: Integrating Large Language Models (LLMs) into complex software systems enables the generation of human-understandable explanations of opaque AI processe

Multimodal Protein Language Models for Enzyme Kinetic Parameters: From Substrate Recognition to Conformational Adaptation

arXiv:2603.12845v2 Announce Type: replace Abstract: Predicting enzyme kinetic parameters quantifies how efficiently an enzyme catalyzes a specific substrate under defined biochemical conditions. Canon

safetyarxiv-cs-cv

Today we open sourced many of OpenAI's monitorability evaluations. We hope that the research community and other model developers can build …

Today we open sourced many of OpenAI's monitorability evaluations. We hope that the research community and other model developers can build upon them and use them to evaluate the monitorability of the

safetysam-altman--x

AgentSOC: A Multi-Layer Agentic AI Framework for Security Operations Automation

arXiv:2604.20134v1 Announce Type: cross Abstract: Security Operations Centers (SOCs) increasingly encounter difficulties in correlating heterogeneous alerts, interpreting multi-stage attack progressio

Ask Only When Needed: Proactive Retrieval from Memory and Skills for Experience-Driven Lifelong Agents

arXiv:2604.20572v1 Announce Type: new Abstract: Online lifelong learning enables agents to accumulate experience across interactions and continually improve on long-horizon tasks. However, existing me

FLOSS: Federated Learning with Opt-Out and Straggler Support

arXiv:2507.23115v2 Announce Type: replace-cross Abstract: Previous work on data privacy in federated learning systems focuses on privacy-preserving operations for data from users who have agreed to sh

LLMs Can Get 'Brain Rot': A Pilot Study on Twitter/X

arXiv:2510.13928v2 Announce Type: replace-cross Abstract: We propose and test the LLM Brain Rot Hypothesis: continual exposure to junk web text induces lasting cognitive decline in large language mode

MGDA-Decoupled: Geometry-Aware Multi-Objective Optimisation for DPO-based LLM Alignment

arXiv:2604.20685v1 Announce Type: new Abstract: Aligning large language models (LLMs) to desirable human values requires balancing multiple, potentially conflicting objectives such as helpfulness, tru

SAMix: Calibrated and Accurate Continual Learning via Sphere-Adaptive Mixup and Neural Collapse

arXiv:2510.15751v2 Announce Type: replace Abstract: While most continual learning methods focus on mitigating forgetting and improving accuracy, they often overlook the critical aspect of network cali

Verification of Machine Unlearning is Fragile

arXiv:2408.00929v2 Announce Type: replace Abstract: As privacy concerns escalate in the realm of machine learning, data owners now have the option to utilize machine unlearning to remove their data fr

BAPO: Boundary-Aware Policy Optimization for Reliable Agentic Search

arXiv:2601.11037v2 Announce Type: replace Abstract: RL-based agentic search enables LLMs to solve complex questions via dynamic planning and external search. While this approach significantly enhances