safety

Batch-Adaptive Causal Annotations

arXiv:2502.10605v3 Announce Type: replace-cross Abstract: Estimating the causal effects of interventions is crucial to policy and decision-making, yet outcome data are often missing or subject to non-

Better with Less: Tackling Heterogeneous Multi-Modal Image Joint Pretraining via Conditioned and Degraded Masked Autoencoder

DGX agent

arXiv:2604.16952v1 Announce Type: new Abstract: Learning robust representations across extremely heterogeneous modalities remains a fundamental challenge in multi-modal vision. As a critical and profo

safetyarxiv-cs-cv

21 Apr 2026

Beyond Overlap Metrics: Rewarding Reasoning and Preferences for Faithful Multi-Role Dialogue Summarization

arXiv:2604.17188v1 Announce Type: new Abstract: Multi-role dialogue summarization requires modeling complex interactions among multiple speakers while preserving role-specific information and factual

Bias-constrained multimodal intelligence for equitable and reliable clinical AI

arXiv:2604.16884v1 Announce Type: new Abstract: The integration of medical imaging and clinical text has enabled the emergence of generalist artificial intelligence (AI) systems for healthcare. Howeve

BIASEDTALES-ML: A Multilingual Dataset for Analyzing Narrative Attribute Distributions in LLM-Generated Stories

arXiv:2604.17008v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly used to generate narrative content, including children's stories, which play an important role in social a

Boltzmann Machine Learning with a Parallel, Persistent Markov chain Monte Carlo method for Estimating Evolutionary Fields and Couplings from a Protein Multiple Sequence Alignment

arXiv:2604.18022v1 Announce Type: cross Abstract: The inverse Potts problem for estimating evolutionary single-site fields and pairwise couplings in homologous protein sequences from their single-site

Bounded Ratio Reinforcement Learning

arXiv:2604.18578v1 Announce Type: new Abstract: Proximal Policy Optimization (PPO) has become the predominant algorithm for on-policy reinforcement learning due to its scalability and empirical robust

BRIDGE the Gap: Mitigating Bias Amplification in Automated Scoring of English Language Learners via Inter-group Data Augmentation

arXiv:2602.23580v2 Announce Type: replace Abstract: In the field of educational assessment, automated scoring systems increasingly rely on deep learning and large language models (LLMs). However, thes

C-GenReg: Training-Free 3D Point Cloud Registration by Multi-View-Consistent Geometry-to-Image Generation with Probabilistic Modalities Fusion

arXiv:2604.16680v1 Announce Type: new Abstract: We introduce C-GenReg, a training-free framework for 3D point cloud registration that leverages the complementary strengths of world-scale generative pr

Can Explicit Physical Feasibility Benefit VLA Learning? An Empirical Study

arXiv:2604.17896v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models map multimodal inputs directly to robot actions and are typically trained through large-scale imitation learning. Wh

CAPC-CG: A Large-Scale, Expert-Directed LLM-Annotated Corpus of Adaptive Policy Communication in China

arXiv:2510.08986v2 Announce Type: replace Abstract: We introduce CAPC-CG, the Chinese Adaptive Policy Communication (Central Government) Corpus, the first open dataset of Chinese policy directives ann

CAPO: Counterfactual Credit Assignment in Sequential Cooperative Teams

arXiv:2604.17693v1 Announce Type: new Abstract: In cooperative teams where agents act in a fixed order and share a single team reward, it is hard to know how much each agent contributed, and harder st

Cat-DPO: Category-Adaptive Safety Alignment

arXiv:2604.17299v1 Announce Type: new Abstract: Aligning large language models with human preferences must balance two competing goals: responding helpfully to legitimate requests and reliably refusin

CCAR: Intrinsic Robustness as an Emergent Geometric Property

arXiv:2604.16861v1 Announce Type: cross Abstract: Standard supervised learning optimizes for predictive accuracy but remains agnostic to the internal geometry of learned features, often yielding repre

CGCMA: Conditionally-Gated Cross-Modal Attention for Event-Conditioned Asynchronous Fusion

arXiv:2604.16411v1 Announce Type: new Abstract: We study asynchronous alignment, a first-class multimodal learning setting in which a dense primary stream must be fused with sporadic external context

Characterizing Model-Native Skills

arXiv:2604.17614v1 Announce Type: cross Abstract: Skills are a natural unit for describing what a language model can do and how its behavior can be changed. However, existing characterizations rely on

Chasing Ghosts: A Simulation-to-Real Olfactory Navigation Stack with Optional Vision Augmentation

arXiv:2602.19577v2 Announce Type: replace Abstract: Autonomous odor source localization remains a challenging problem for aerial robots due to turbulent airflow, sparse and delayed sensory signals, an

safetyarxiv-cs-ro

Closing the Modality Reasoning Gap for Speech Large Language Models

arXiv:2601.05543v2 Announce Type: replace Abstract: Although Speech Large Language Models have achieved notable progress, a substantial modality reasoning gap remains: their reasoning performance on s

COFFAIL: A Dataset of Successful and Anomalous Robot Skill Executions in the Context of Coffee Preparation

arXiv:2604.18236v1 Announce Type: new Abstract: In the context of robot learning for manipulation, curated datasets are an important resource for advancing the state of the art; however, available dat

safetyarxiv-cs-ro

Cognitive Policy-Driven LLM for Diagnosis and Intervention of Cognitive Distortions in Emotional Support Conversation

arXiv:2604.17178v1 Announce Type: new Abstract: Emotional Support Conversation (ESC) plays a critical role in mental health assistance by providing accessible psychological support in real-world appli

Composed Vision-Language Retrieval for Skin Cancer Case Search via Joint Alignment of Global and Local Representations

arXiv:2603.09108v2 Announce Type: replace Abstract: Medical image retrieval aims to identify clinically relevant lesion cases to support diagnostic decision making, education, and quality control. In

Contact-Rich Robotic Assembly in Construction via Diffusion Policy Learning

arXiv:2511.17774v3 Announce Type: replace Abstract: Fabrication uncertainty arising from tolerance accumulation, material imperfection, and positioning errors remains a critical barrier to automated r

safetyarxiv-cs-ro

Continual Safety Alignment via Gradient-Based Sample Selection

arXiv:2604.17215v1 Announce Type: new Abstract: Large language models require continuous adaptation to new tasks while preserving safety alignment. However, fine-tuning on even benign data often compr

Contrastive Analysis of Linguistic Representations in Large Language Model Outputs through Structured Synthetic Data Generation and Abstracted N-gram Associations

arXiv:2604.17398v1 Announce Type: new Abstract: We present a methodological framework to discover linguistic and discursive patterns associated to different social groups through contrastive synthetic

Cooperative Coevolution versus Monolithic Evolutionary Search for Semi-Supervised Tabular Classification

arXiv:2604.16412v1 Announce Type: cross Abstract: This paper studies semi-supervised tabular classification in the extreme low-label regime using lightweight base learners. The paper proposes a cooper

Correction and Corruption: A Two-Rate View of Error Flow in LLM Protocols

arXiv:2604.18245v1 Announce Type: new Abstract: Large language models are increasingly deployed as protocols: structured multi-call procedures that spend additional computation to transform a baseline

COSEARCH: Joint Training of Reasoning and Document Ranking via Reinforcement Learning for Agentic Search

arXiv:2604.17555v1 Announce Type: cross Abstract: Agentic search -- the task of training agents that iteratively reason, issue queries, and synthesize retrieved information to answer complex questions

CRISP: Compressing Redundancy in Chain-of-Thought via Intrinsic Saliency Pruning

arXiv:2604.17297v1 Announce Type: new Abstract: Long Chain-of-Thought (CoT) reasoning is pivotal for the success of recent reasoning models but suffers from high computational overhead and latency. Wh

Cross-Modal Attention Analysis and Optimization in Vision-Language Models: A Study on Visual Reliability

arXiv:2604.17217v1 Announce Type: new Abstract: Vision-Language Models (VLMs) achieve strong cross-modal performance, yet recent evidence suggests they over-rely on textual descriptions while under-ut

CrossFlowDG: Bridging the Modality Gap with Cross-modal Flow Matching for Domain Generalization

arXiv:2604.16892v1 Announce Type: new Abstract: Domain generalization (DG) aims to maintain performance under domain shift, which in computer vision appears primarily as stylistic variations that caus

Culture-Aware Humorous Captioning: Multimodal Humor Generation across Cultural Contexts

arXiv:2604.18091v1 Announce Type: new Abstract: Recent multimodal large language models have shown promising ability in generating humorous captions for images, yet they still lack stable control over

Debate as Reward: A Multi-Agent Reward System for Scientific Ideation via RL Post-Training

arXiv:2604.16723v1 Announce Type: cross Abstract: Large Language Models (LLMs) have demonstrated potential in automating scientific ideation, yet current approaches relying on iterative prompting or c

Decoding AI Tutor Effects for Educational Measurement: Temporal, Multi-Outcome, and Behavior-Cognitive Analysis

arXiv:2604.16366v1 Announce Type: cross Abstract: Artificial intelligence (AI) tutors have become increasingly popular in learning environments. In this study, we propose an AI agent prototype framewo

Decoupling the Effect of Chain-of-Thought Reasoning: A Human Label Variation Perspective

arXiv:2601.03154v2 Announce Type: replace Abstract: Reasoning-tuned LLMs utilizing long Chain-of-Thought (CoT) excel at single-answer tasks, yet their ability to model Human Label Variation--which req

Deep learning based Non-Rigid Volume-to-Surface Registration for Brain Shift compensation Using Point Cloud

arXiv:2604.17389v1 Announce Type: new Abstract: Soft-tissue deformation remains a major limitation in image-guided neurosurgery, where intra-operative anatomy can deviate substantially from pre-operat

DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models

arXiv:2511.15669v2 Announce Type: replace Abstract: Does Chain-of-Thought (CoT) reasoning genuinely improve Vision-Language-Action (VLA) models, or does it merely add overhead? Existing CoT-VLA system

Demystifying the unreasonable effectiveness of online alignment methods

arXiv:2604.17207v1 Announce Type: cross Abstract: Iterative alignment methods based on purely greedy updates are remarkably effective in practice, yet existing theoretical guarantees of (O(log T)) KL-

Detecting Alarming Student Verbal Responses using Text and Audio Classifier

arXiv:2604.16717v1 Announce Type: new Abstract: This paper addresses a critical safety gap in the use Automated Verbal Response Scoring (AVRS). We present a novel hybrid framework for troubled student

DFedReweighting: A Unified Framework for Objective-Oriented Reweighting in Decentralized Federated Learning

arXiv:2512.12022v2 Announce Type: replace Abstract: Decentralized federated learning (DFL) has emerged as a promising paradigm that enables multiple clients to collaboratively train machine learning m

Diagnosing LLM-based Rerankers in Cold-Start Recommender Systems: Coverage, Exposure and Practical Mitigations

arXiv:2604.16318v1 Announce Type: cross Abstract: Large language models (LLMs) and cross-encoder rerankers have gained attention for improving recommender systems, particularly in cold-start scenarios

DiffCoT: Diffusion-styled Chain-of-Thought Reasoning in LLMs

arXiv:2601.03559v2 Announce Type: replace Abstract: Chain-of-Thought (CoT) reasoning improves multi-step mathematical problem solving in large language models but remains vulnerable to exposure bias a

Differential Privacy in Two-Layer Networks: How DP-SGD Harms Fairness and Robustness

arXiv:2603.04881v2 Announce Type: replace Abstract: Differentially private learning is essential for training models on sensitive data, but empirical studies consistently show that it can degrade perf

Distributional Off-Policy Evaluation with Deep Quantile Process Regression

arXiv:2604.18143v1 Announce Type: cross Abstract: This paper investigates the off-policy evaluation (OPE) problem from a distributional perspective. Rather than focusing solely on the expectation of t

Diverse Dictionary Learning

arXiv:2604.17568v1 Announce Type: new Abstract: Given only observational data X = g(Z), where both the latent variables Z and the generating process g are unknown, recovering Z is ill-posed without ad

DMax: Aggressive Parallel Decoding for dLLMs

arXiv:2604.08302v2 Announce Type: replace Abstract: We present DMax, a new paradigm for efficient diffusion language models (dLLMs). It mitigates error accumulation in parallel decoding, enabling aggr

Do LLMs Use Cultural Knowledge Without Being Told? A Multilingual Evaluation of Implicit Pragmatic Adaptation

arXiv:2604.17718v1 Announce Type: new Abstract: Many benchmarks show that large language models can answer direct questions about culture. We study a different question: do they also change how they s

Does 'Do Differentiable Simulators Give Better Policy Gradients?'' Give Better Policy Gradients?

arXiv:2604.18161v1 Announce Type: new Abstract: In policy gradient reinforcement learning, access to a differentiable model enables 1st-order gradient estimation that accelerates learning compared to

Does Welsh media need a review? Detecting bias in Nation.Cymru's political reporting

arXiv:2604.17628v1 Announce Type: new Abstract: Wales' political landscape has been marked by growing accusations of bias in Welsh media. This paper takes the first computational step toward testing t