Search: “safety”

Beyond Match Maximization and Fairness: Retention-Optimized Two-Sided Matching

arXiv:2602.15752v2 Announce Type: replace Abstract: On two-sided matching platforms such as online dating and recruiting, recommendation algorithms often aim to maximize the total number of matches. H

Breaking Lock-In: Preserving Steerability under Low-Data VLA Post-Training

DGX agent

arXiv:2604.23121v1 Announce Type: cross Abstract: Have you ever post-trained a generalist vision-language-action (VLA) policy on a small demonstration dataset, only to find that it stops responding to

safetyarxiv-cs-cv

28 Apr 2026

Bridging Reasoning and Action: Hybrid LLM-RL Framework for Efficient Cross-Domain Task-Oriented Dialogue

arXiv:2604.23345v1 Announce Type: new Abstract: Cross-domain task-oriented dialogue requires reasoning over implicit and explicit feasibility constraints while planning long-horizon, multi-turn action

BVI-Mamba: Video Enhancement Using a Visual State-Space Model for Low-Light and Underwater Environments

arXiv:2604.23655v1 Announce Type: new Abstract: Videos captured in low-light and underwater conditions often suffer from distortions such as noise, low contrast, color imbalance, and blur. These issue

CA-IDD: Cross-Attention Guided Identity-Conditional Diffusion for Identity-Consistent Face Swapping

arXiv:2604.24493v1 Announce Type: new Abstract: Face swapping aims to optimize realistic facial image generation by leveraging the identity of a source face onto a target face while preserving pose, e

Can Compact Language Models Search Like Agents? Distillation-Guided Policy Optimization for Preserving Agentic RAG Capabilities

arXiv:2508.20324v4 Announce Type: replace Abstract: Reinforcement Learning has emerged as a dominant post-training approach to elicit agentic RAG behaviors such as search and planning from language mo

CASP: Support-Aware Offline Policy Selection for Two-Stage Recommender Systems

arXiv:2604.23022v1 Announce Type: cross Abstract: Two-stage recommender systems first choose a candidate generator and then rank items within the generated set. Because the generator decides which ite

CODA: Coordination via On-Policy Diffusion for Multi-Agent Offline Reinforcement Learning

arXiv:2604.23308v1 Announce Type: new Abstract: Offline multi-agent reinforcement learning (MARL) enables policy learning from fixed datasets, but is prone to coordination failure: agents trained on s

CoFi-PGMA: Counterfactual Policy Gradients under Filtered Feedback for Multi-Agent LLMs

arXiv:2604.22785v1 Announce Type: new Abstract: Large language model (LLM) deployments increasingly rely on multi-agent architectures in which multiple models either compete through routing mechanisms

COMO: Closed-Loop Optical Molecule Recognition with Minimum Risk Training

arXiv:2604.23546v1 Announce Type: cross Abstract: Optical chemical structure recognition (OCSR) translates molecular images into machine-readable representations like SMILES strings or molecular graph

Complex SGD and Directional Bias in Reproducing Kernel Hilbert Spaces

arXiv:2604.23017v1 Announce Type: new Abstract: Stochastic Gradient Descent (SGD) is a known stochastic iterative method popular for large-scale convex optimization problems due to its simple implemen

Conditional Imputation for Within-Modality Missingness in Multi-Modal Federated Learning

arXiv:2604.23112v1 Announce Type: new Abstract: Multimodal Federated Learning (MMFL) enables privacy-preserving collaborative training, but real-world clinical applications often suffer from within-mo

Conflict-Aware Harmonized Rotational Gradient for Multiscale Kinetic Regimes

arXiv:2604.24745v1 Announce Type: new Abstract: In this paper, we propose a harmonized rotational gradient method, termed HRGrad, for simultaneously tackling multiscale time-dependent kinetic problems

ConsDreamer: Advancing Multi-View Consistency for Zero-Shot Text-to-3D Generation

arXiv:2504.02316v4 Announce Type: replace-cross Abstract: Recent advances in zero-shot text-to-3D generation have revolutionized 3D content creation by enabling direct synthesis from textual descripti

CT-Guided Spatially-varying Regularization for Voxel-Wise Deformable Whole-Body PET Registration

arXiv:2604.22905v1 Announce Type: cross Abstract: Whole-body Positron Emission Tomography (PET) registration is essential for multi-parametric tumor characterization and assessment of metastatic disea

CURE-Med: Curriculum-Informed Reinforcement Learning for Multilingual Medical Reasoning

arXiv:2601.13262v2 Announce Type: replace Abstract: While large language models (LLMs) have shown to perform well on monolingual mathematical and commonsense reasoning, they remain unreliable for mult

Data-efficient Targeted Token-level Preference Optimization for LLM-based Text-to-Speech

arXiv:2510.05799v2 Announce Type: replace-cross Abstract: Aligning text-to-speech (TTS) system outputs with human feedback through preference optimization has been shown to effectively improve the rob

DextER: Language-driven Dexterous Grasp Generation with Embodied Reasoning

arXiv:2601.16046v2 Announce Type: replace-cross Abstract: Language-driven dexterous grasp generation requires the models to understand task semantics, 3D geometry, and complex hand-object interactions

DLM: Unified Decision Language Models for Offline Multi-Agent Sequential Decision Making

arXiv:2604.23557v1 Announce Type: cross Abstract: Building scalable and reusable multi-agent decision policies from offline datasets remains a challenge in offline multi-agent reinforcement learning (

Do Synthetic Trajectories Reflect Real Reward Hacking? A Systematic Study on Monitoring In-the-Wild Hacking in Code Generation

arXiv:2604.23488v1 Announce Type: new Abstract: Reward hacking in code generation, where models exploit evaluation loopholes to obtain full reward without correctly solving the tasks, poses a critical

Do Transaction-Level and Actor-Level AML Queues Agree? An Empirical Evaluation of Granularity Effects on the Elliptic++ Graph

arXiv:2604.23494v1 Announce Type: new Abstract: Graph-based anti-money laundering (AML) systems on blockchain networks can score suspicious activity at two granularity levels -- transactions or actor

DPEPO: Diverse Parallel Exploration Policy Optimization for LLM-based Agents

arXiv:2604.24320v1 Announce Type: new Abstract: Large language model (LLM) agents that follow the sequential 'reason-then-act' paradigm have achieved superior performance in many complex tasks.However

DPRM: A Plug-in Doob h transform-induced Token-Ordering Module for Diffusion Language Models

arXiv:2604.24357v1 Announce Type: cross Abstract: Diffusion language models generate without a fixed left-to-right order, making token ordering a central algorithmic choice: which tokens should be rev

DVPO: Distributional Value Modeling-based Policy Optimization for LLM Post-Training

arXiv:2512.03847v2 Announce Type: replace-cross Abstract: Reinforcement learning (RL) has shown strong performance in LLM post-training, but real-world deployment often involves noisy or incomplete su

EAD-Net: Emotion-Aware Talking Head Generation with Spatial Refinement and Temporal Coherence

arXiv:2604.23325v1 Announce Type: cross Abstract: Emotionally talking head video generation aims to generate expressive portrait videos with accurate lip synchronization and emotional facial expressio

Efficient Rationale-based Retrieval: On-policy Distillation from Generative Rerankers based on JEPA

arXiv:2604.23336v1 Announce Type: cross Abstract: Unlike traditional fact-based retrieval, rationale-based retrieval typically necessitates cross-encoding of query-document pairs using large language

EL3DD: Extended Latent 3D Diffusion for Language Conditioned Multitask Manipulation

arXiv:2511.13312v2 Announce Type: replace-cross Abstract: Acting in human environments is a crucial capability for general-purpose robots, necessitating a robust understanding of natural language and

EU countries and lawmakers reach an impasse on a deal watering down the EU's AI Act due to some parties seeking exemptions for already regulated industries (Foo Yun Chee/Reuters)

Foo Yun Chee / Reuters: EU countries and lawmakers reach an impasse on a deal watering down the EU's AI Act due to some parties seeking exemptions for already regulated industries — EU countries and E

safetytechmeme

Evaluating Language Models' Evaluations of Games

arXiv:2510.10930v2 Announce Type: replace-cross Abstract: Reasoning is not just about solving problems -- it is also about evaluating which problems are worth solving at all. Evaluations of artificial

Explanation Quality Assessment as Ranking with Listwise Rewards

arXiv:2604.24176v1 Announce Type: new Abstract: We reformulate explanation quality assessment as a ranking problem rather than a generation problem. Instead of optimizing models to produce a single 'b

Extreme bandits

arXiv:2604.24545v1 Announce Type: cross Abstract: In many areas of medicine, security, and life sciences, we want to allocate limited resources to different sources in order to detect extreme values.

Failure-Centered Runtime Evaluation for Deployed Trilingual Public-Space Agents

arXiv:2604.23990v1 Announce Type: new Abstract: This paper presents PSA-Eval, a failure-centered runtime evaluation framework for deployed trilingual public-space agents. The central claim is that, wh

Federated Cross-Modal Retrieval with Missing Modalities via Semantic Routing and Adapter Personalization

arXiv:2604.22885v1 Announce Type: cross Abstract: Federated cross-modal retrieval faces severe challenges from heterogeneous client data, particularly non-IID semantic distributions and missing modali

Fine-R1: Make Multi-modal LLMs Excel in Fine-Grained Visual Recognition by Chain-of-Thought Reasoning

arXiv:2602.07605v3 Announce Type: replace-cross Abstract: Any entity in the visual world can be hierarchically grouped based on shared characteristics and mapped to fine-grained sub-categories. While

FinGround: Detecting and Grounding Financial Hallucinations via Atomic Claim Verification

arXiv:2604.23588v1 Announce Type: new Abstract: Financial AI systems must produce answers grounded in specific regulatory filings, yet current LLMs fabricate metrics, invent citations, and miscalculat

From what I can tell, Elon Musk just stepped on his own toes at the trial, making it all about him instead of the promises Altman and Brockm…

From what I can tell, Elon Musk just stepped on his own toes at the trial, making it all about him instead of the promises Altman and Brockman broke. OpenAI will slay his ego on cross. He should have

safetygary-marcus--x

GAMED.AI: A Hierarchical Multi-Agent Framework for Automated Educational Game Generation

arXiv:2604.23947v1 Announce Type: new Abstract: We introduce GameDAI, a hierarchical multi-agent framework that transforms instructor-provided questions into fully playable, pedagogically grounded edu

Google, 2001: Don’t Be Evil Google, 2026: How much does mass surveillance pay?

This post by AI researcher Gary Marcus contrasts Google's original 'Don't Be Evil' corporate motto from 2001 with a critical question about the company's 2026 practices, likely arguing that Google has

safetygary-marcus--x

Hierarchical Prototype-based Domain Priors for Multiple Instance Learning in Multimodal Histopathology Analysis

arXiv:2604.23982v1 Announce Type: new Abstract: Digital pathology has fundamentally altered diagnostic workflows by enabling the computational analysis of gigapixel Whole Slide Images (WSIs), yet effe

Hindsight Preference Optimization for Financial Time Series Advisory

arXiv:2604.23988v1 Announce Type: cross Abstract: Time series models predict numbers; decision-makers need advisory -- directional signals with reasoning, actionable suggestions, and risk management.

Humanoid Whole-Body Badminton via Multi-Stage Reinforcement Learning

arXiv:2511.11218v3 Announce Type: replace Abstract: Humanoid robots have demonstrated strong capabilities for interacting with static scenes across locomotion and manipulation, yet dynamic real-world

safetyarxiv-cs-ro

In-Sync: Adaptation of Speech Aware Large Language Models for ASR with Word Level Timestamp Predictions

arXiv:2604.22817v1 Announce Type: cross Abstract: Recent advances in speech-aware language models have coupled strong acoustic encoders with large language models, enabling systems that move beyond tr

In this informal poll, only a fifth of all coders and software engineers have gone over full time to vibe coding. Most use vibe coding; few …

In this informal poll, only a fifth of all coders and software engineers have gone over full time to vibe coding. Most use vibe coding; few use it exclusively. Most still do some hand coding. Amateurs

safetygary-marcus--x

InCoM: Intent-Driven Perception and Structured Coordination for Mobile Manipulation

arXiv:2602.23024v2 Announce Type: replace Abstract: Mobile manipulation is a fundamental capability for general-purpose robotic agents, requiring both coordinated control of the mobile base and manipu

safetyarxiv-cs-ro

Institutions for the Post-Scarcity of Judgment

arXiv:2604.22966v1 Announce Type: cross Abstract: Each major technological revolution inverts a particular scarcity and rebuilds institutions around the shift. The near-consensus diagnosis of the AI r

Intervention-Aware Multiscale Representation Learning from Imaging Phenomics and Perturbation Transcriptomics

arXiv:2604.22832v1 Announce Type: cross Abstract: Microscopy-based phenotypic profiling is scalable for drug discovery but lacks the mechanistic depth of transcriptomics, which remains costly and scar

IRIS: Interleaved Reinforcement with Incremental Staged Curriculum for Cross-Lingual Mathematical Reasoning

arXiv:2604.24114v1 Announce Type: new Abstract: Curriculum learning helps language models tackle complex reasoning by gradually increasing task difficulty. However, it often fails to generate consiste

Is Chain-of-Thought Reasoning of LLMs a Mirage? A Data Distribution Lens

arXiv:2508.01191v5 Announce Type: replace Abstract: Chain-of-Thought (CoT) prompting has been shown to be effective in eliciting structured reasoning (i.e., CoT reasoning) from large language models (