Search: “safety”

Analytic Concept-Centric Memory for Agentic Embodied Manipulation

arXiv:2606.29774v1 Announce Type: new Abstract: Long-horizon embodied manipulation requires agents to remember persistent objects, track changing scene states, and reuse prior interaction knowledge. H

Analyzing Defensive Misdirection Against Model-Guided Automated Attacks on Agentic AI Systems

DGX agent

arXiv:2606.20470v2 Announce Type: replace-cross Abstract: Agentic AI systems increasingly rely on language-model components to interpret instructions, process external data, invoke tools, and coordina

safetyarxiv-cs-ai

30 Jun 2026

Anomaly detection using dynamic thresholds and two-year-long alerts in Cloud Monitoring

Choosing the threshold of an alert policy can be a headache. You have to analyze historical data, aggregate it into semantically meaningful time series, and choose a threshold that matters. If the wor

safetygoogle-cloud-ai

AnyBody: Free-Form Whole-Body Humanoid Control from Arbitrary Keypoint Guidance

arXiv:2606.29209v1 Announce Type: cross Abstract: We present AnyBody, a unified whole-body humanoid controller driven by an arbitrary subset of body keypoints chosen at deploy time. Prior physics-base

Are Humans Evolved Instruction Followers? An Underlying Inductive Bias Enables Rapid Instructed Task Learning

arXiv:2606.29792v1 Announce Type: new Abstract: Human adults can often perform a novel task correctly on the first attempt after only receiving verbal or written instructions. This rapid instructed ta

Are We Measuring Strategy or Phrasing? The Gap Between Surface- and Approach-Level Diversity in LLM Math Reasoning

arXiv:2606.29985v1 Announce Type: new Abstract: Diversity in LLM mathematical reasoning is critical for exploration, but common diversity metrics mostly capture surface-level variation rather than dif

Are Whitepaper Claims Reflected in Market Structure? A Contamination-Aware Pipeline and a Power-Limited Null

arXiv:2601.20336v5 Announce Type: replace-cross Abstract: Do the functional narratives in cryptocurrency whitepapers correspond to how their tokens behave in markets? We develop a content-verified, co

ARKD: Adaptive Reinforcement Learning-Guided Bidirectional KL Divergence Distillation for Text Generation

arXiv:2606.29869v1 Announce Type: cross Abstract: Knowledge distillation (KD) is a key technique for compressing Large Language Models (LLMs), yet methods relying on a single KL objective often fail t

Bandwidth Selection in Kernel Density Estimation for Model Calibration

arXiv:2606.29925v1 Announce Type: new Abstract: As deep learning models are increasingly deployed in high-stakes applications, providing well-calibrated uncertainty estimates has become as critical as

Be Faithful When Response: Returning Fluent and Grounded Answers for Vision-Language Models Reinforcement Learning

arXiv:2606.29984v1 Announce Type: new Abstract: Reinforcement Learning (RL) is an important paradigm for improving the reasoning capabilities of Vision-Language Models (VLMs). However, directly applyi

Behavior Prompting Policy: Demonstrations as Prompts for Manipulation

arXiv:2606.30457v1 Announce Type: new Abstract: We study behavior prompting, a paradigm that enables robots to perform new tasks at inference time given a single human demonstration, which we call a b

Behavior Uncloning: Distilling Mode Redirection into Policy Weights without Inference-Time Steering

arXiv:2606.29201v1 Announce Type: cross Abstract: Behavior-cloned policies often learn multiple behavior modes from demonstration datasets, including modes that are unsafe or otherwise undesired at de

Beyond Backscatter: AlphaEarth Land-Cover Priors for Rapid SAR Flood Segmentation Across Foundation Backbones

arXiv:2606.29134v1 Announce Type: new Abstract: Rapid flood mapping is critical for emergency response, yet optical imagery is often unusable during major flooding and single-temporal SAR is ambiguous

BrainJanus: A Unified Model for Understanding and Generation across Brain, Vision, and Language

arXiv:2606.30319v1 Announce Type: new Abstract: Modeling the bidirectional correspondence between external sensory stimuli and internal neural activity has emerged as a critical frontier in neuroscien

Bridgewater, one of the worlds largest hedge funds, a Tinker customer talks through how they've carefully fine-tuned a model focused on what…

safetysoumith-chintala--x

Bridgewater, one of the worlds largest hedge funds, a Tinker customer talks through how they've carefully fine-tuned a model focused on what makes interesting financial news. Their fine-tuned model is

BTI-Net: Bidirectional Decoder-Level Task Interaction via Uncertainty-Aware Gating for Multi-Task Medical Image Analysis

arXiv:2606.29102v1 Announce Type: cross Abstract: Jointly learning to segment and classify medical images demands cross-task synergy, yet encoder-sharing architectures limit decoder reconstruction to

Building Multi-Task Agentic LLMs via Two-Phase Distillation

arXiv:2606.30044v1 Announce Type: new Abstract: A key step toward artificial general intelligence is to train models that can perform multiple tasks. In this paper, we study how to build such models b

BV-Blend: Uncertainty-Weighted Historical Baselines for Stable Critic-Free RL with Verifiable Rewards

arXiv:2606.28707v1 Announce Type: new Abstract: Critic-free reinforcement learning with verifiable rewards (RLVR), exemplified by Group Relative Policy Optimization (GRPO), avoids training a value fun

CAMI: Cost-Aware Agent-Guided Multi-Indexing for Semantic Retrieval

arXiv:2606.28365v1 Announce Type: cross Abstract: RAG ingestion pipelines frequently augment search corpus index with semantic enrichment indices (e.g., synthetic queries or summaries generated from c

Can MLLMs Critique Like Humans? Evaluating Open-Ended Aesthetic Reasoning in Multimodal Large Language Models

arXiv:2606.29689v1 Announce Type: new Abstract: Open-ended aesthetic critique is a challenge for multimodal large language models (MLLMs): unlike multiple-choice aesthetic benchmarks, it has no single

Categorizing Mathematical Concepts with LLM Voting Ensembles in Mathswitch

arXiv:2606.28815v1 Announce Type: cross Abstract: Mathswitch is an open-source project that imports mathematical concept records from sources such as Wikidata, Wikipedia, MathWorld, Encyclopedia of Ma

Chronos: A Physics-Informed Full-History Framework for Non-Markovian Long-Horizon Manipulation

arXiv:2606.30318v1 Announce Type: new Abstract: General-purpose robot policies should be modeled as dynamical systems, yet many VLA and generative imitation policies still rely on present observations

Clearer Sight, Fewer Lies: Oriented Pickup Preference Optimization for Multimodal Hallucination Mitigation

arXiv:2606.29805v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) are prone to hallucination as their generation preferences are insufficiently calibrated to visual evidence, ca

CMTFormer: Marrying Transformer with Hierarchical Information Interaction for RGB-Event Object Detection

arXiv:2606.29136v1 Announce Type: cross Abstract: Event cameras capture sparse brightness changes with high temporal resolution and high dynamic range, compensating for the deficiencies of the convent

Complementary RL: Towards Efficient Experience-Driven Agent Learning

arXiv:2603.17621v2 Announce Type: replace-cross Abstract: Reinforcement Learning (RL) has emerged as a powerful paradigm for training LLM-based agents, yet remains limited by low sample efficiency, st

ConCent: Contact-Centric Real-to-Sim-to-Real Learning from One Demonstration

arXiv:2606.30268v1 Announce Type: new Abstract: Sim-to-real policy transfer -- deploying policies trained in simulation in the real world -- is a promising paradigm for scaling robot manipulation with

Consistency as Inductive Bias: Learning Cross-View Invariance for Robust Multimodal Reasoning

arXiv:2606.29812v1 Announce Type: new Abstract: Inductive biases steer learning toward generalizable solutions by encoding task structure. In this work, we identify a crucial missing bias in MLLMs: cr

CORE: Common Outcome Regularities from Action-Free Visual Demonstrations for Robot Manipulation

arXiv:2606.29517v1 Announce Type: new Abstract: Robot imitation learning often relies on costly robot demonstrations, while abundant action-free visual demonstrations, such as human videos, are diffic

CRAFT: Counterfactual Credit Assignment from Free Sibling Rollouts for Self-Distilled Agentic Reinforcement Learning

arXiv:2606.29476v1 Announce Type: cross Abstract: Self-distilled agentic reinforcement learning augments trajectory-level reward with a token-level distillation loss, using as its teacher the same pol

Critical Interval MSE: Toward Reliable Offline Validation for Robot Manipulation Policies

arXiv:2606.29898v1 Announce Type: cross Abstract: Real-world evaluation is the gold standard for robot policies because it tests them against the physical conditions and deployment challenges they are

Data-Efficient Multimodal Alignment for Histopathology-based Molecular Prediction

arXiv:2606.29949v1 Announce Type: cross Abstract: H&E-stained whole-slide images offer cohort-scale availability and rich spatial context but lack molecular specificity, whereas bulk RNA-seq provides

Delayed Bidirectional Alignment via Disentangled Audio Semantics for Audio-Visual Segmentation

arXiv:2512.20117v2 Announce Type: replace Abstract: Audio-Visual Segmentation (AVS) aims to localize sound-producing objects at the pixel level by integrating auditory and visual cues. However, existi

Deterministic Decisions for High-Stakes AI. A Zero-Egress Pipeline with the Deployability of RAG and the Accuracy of Machine Learning

arXiv:2606.29280v1 Announce Type: cross Abstract: We identify intervention bias as a previously unquantified failure mode of zero-shot large-language-model (LLM) educational advisory agents: without t

Diffusion Fine-tuning with Rewarded Moment Matching Distillation

arXiv:2606.30414v1 Announce Type: new Abstract: Distillation and Reinforcement Learning (RL) fine-tuning are the primary pillars of diffusion post-training. While traditionally studied in isolation, t

Distribution Matching Variational AutoEncoder

arXiv:2512.07778v2 Announce Type: replace Abstract: Most visual generative models compress images into a latent space before applying diffusion or autoregressive modelling. Yet, existing approaches su

Distributionally Robust Reinforcement Learning with Human Feedback

arXiv:2503.00539v2 Announce Type: replace-cross Abstract: Reinforcement learning from human feedback (RLHF) has evolved to be one of the main methods for fine-tuning large language models (LLMs). Howe

Do Models Read What They Write? Causal Registers in Scratchpad Reasoning

arXiv:2606.29522v1 Announce Type: cross Abstract: A central hope behind process supervision is that models can expose intermediate variables that matter for their later behavior. For this to help with

DOPD: Dual On-policy Distillation

arXiv:2606.30626v1 Announce Type: new Abstract: On-policy distillation (OPD) offers superior capacity transfer by supervising student-sampled trajectories with dense token-level signals. To furnish hi

DRIFT: Difficulty Routing Self-DIstillation with Rhythm-Gated Exploration and Success BuFfer Training

arXiv:2606.30345v1 Announce Type: cross Abstract: Enabling large language models to achieve stable self-improvement without external expert supervision remains a central challenge in complex reasoning

DrivenMorph: Bridging Attention Mechanism and Variational Image Registration via Difference Modeling

arXiv:2606.30183v1 Announce Type: new Abstract: Medical image registration benefits significantly from deep learning, yet existing approaches often lack physical explainability and fine-grained deform

DTI: Dynamic Trajectory Initialization for Generative Face Video Super-Resolution

arXiv:2606.29198v1 Announce Type: new Abstract: As the most perceptually powerful Face Video Super-Resolution (FVSR) method, existing works in Generative FVSR (GFVSR) mainly exploit the generative pri

Dual-Flow Reinforcement Learning with State-Aware Exploration

arXiv:2606.29820v1 Announce Type: cross Abstract: In complex continuous-control reinforcement learning tasks, multimodal optimal actions often coincide with uncertain, multimodal return distributions,

DyGnROLE: Asymmetric Pretraining for Edge Classification on Dynamic Graphs

arXiv:2602.23135v2 Announce Type: replace-cross Abstract: Edge classification on directed dynamic graphs requires modeling interactions between source and destination nodes exhibiting asymmetrical beh

EntroRouter: Learning Efficient Model Routing via Entropy Regulation

arXiv:2606.29424v1 Announce Type: new Abstract: Model routing balances solution accuracy and computational cost by selecting among models of varying capabilities. While recent multi-round frameworks i

EPIC-EuroParl-UdS: Information-Theoretic Perspectives on Translation and Interpreting

arXiv:2603.09785v3 Announce Type: replace Abstract: This paper introduces an updated and combined version of the bidirectional English-German EPIC-UdS (spoken) and EuroParl-UdS (written) corpora conta

EraseLoRA: MLLM-Driven Foreground Exclusion and Background Subtype Aggregation for Dataset-Free Object Removal

arXiv:2512.21545v2 Announce Type: replace Abstract: Object removal must prevent the masked target from reappearing and reconstruct the occluded background with structural and contextual fidelity, rath

Estimating Grammatical Gender Directions in Contextual Embeddings under Controlled and Natural Contexts

arXiv:2606.30152v1 Announce Type: cross Abstract: Contextual language models conflate grammatical gender and social semantic bias in gendered languages such as Spanish. Existing gender debiasing appro

Evidence-Based Text-Conditioned 3D CT Synthesis for Ovarian Cancer

arXiv:2606.28980v1 Announce Type: cross Abstract: Ovarian cancer is frequently diagnosed at an advanced stage, making preoperative contrast-enhanced computed tomography (CT) central to staging and sur