Search: “safety”

Beyond Parallel Tracking: Interactive Multi-Feature Fusion Drives Semantic Reconstruction from Non-invasive Brain Recordings

arXiv:2607.12071v1 Announce Type: new Abstract: Continuous semantic reconstruction from non-invasive neural recordings remains limited by the representational mismatch between semantic feature spaces

Safety

Calculating Mutual Information between a Reward Maximizer and its Environment

DGX agent

arXiv:2602.12963v2 Announce Type: replace Abstract: An important question in the field of AI is the extent to which successful behaviour requires an internal representation of the world. In this work,

safetyarxiv-cs-ai

15 Jul 2026

Calibrated Selective Prediction Using Deep Ensembles for ROI-Based Thyroid Nodule Ultrasound Classification Under Dataset Shift: A Retrospective Evaluation

arXiv:2607.12075v1 Announce Type: cross Abstract: Background: Deep learning models can classify thyroid nodules on ultrasound, but reliable clinical decision support also requires calibrated probabili

Calibration-First Reward-Component Auditing for Reinforcement Learning Control in Smart Greenhouses

arXiv:2607.11959v1 Announce Type: new Abstract: Greenhouse reinforcement learning can test climate-control ideas at a speed and scale that is difficult to achieve with crop experiments alone. For smar

Can LLMs Write Reliable Rubrics? A Meta-Evaluation for Experiment Reproduction

arXiv:2607.12835v1 Announce Type: new Abstract: Rubric-based evaluation is a promising approach for assessing open-ended outputs from LLM-based research agents, particularly in paper reproduction, whe

CASHEW: Stabilizing Multimodal Reasoning via Iterative Trajectory Aggregation

arXiv:2601.08010v3 Announce Type: replace Abstract: Vision-language models achieve strong performance across a wide range of multimodal understanding and reasoning tasks, yet their multi-step reasonin

CGRL: Concept-Guided Pruning and Representation Learning for Whole-Slide Image Classification

arXiv:2607.12556v1 Announce Type: new Abstract: Weakly supervised whole-slide image (WSI) classification is widely used in computational pathology because slide-level labels are easier to obtain than

ChunkFlow: Towards Continuity-Consistent Chunked Policy Learning

arXiv:2607.12992v1 Announce Type: new Abstract: Vision-language action (VLA) models increasingly adopt chunked action heads to satisfy real-time constraints; however, this introduces boundary jitter:

CityBehavEx: A Scalable and Empirically Validated LLM-Assisted Urban Simulation Platform

arXiv:2607.12086v1 Announce Type: new Abstract: Recent LLM-based multi-agent urban simulators can generate semantically rich city routines, but they remain costly to scale and are often weakly validat

Compos3D: Interactive Part-Based Composition for Creative Control in Generative 3D Models

arXiv:2607.12193v1 Announce Type: cross Abstract: While generative AI has unlocked new opportunities for 3D content creation, current workflows often rely on multiple regenerations, which provides lim

Continuous Policy and Value Iteration for Stochastic Control Problems and Its Convergence

arXiv:2506.08121v2 Announce Type: replace-cross Abstract: We introduce a continuous policy-value iteration algorithm where the approximations of the value function of a stochastic control problem and

DenseReward: Dense Reward Learning via Failure Synthesis for Robotic Manipulation

arXiv:2607.13033v1 Announce Type: new Abstract: Reinforcement learning holds great promise for improving robot policies beyond the limits of imitation learning. However, its practical adoption remains

Deployable Human Preference Alignment in Robotics: Learning Representative Rewards from Diverse Human Preferences

arXiv:2607.12466v1 Announce Type: new Abstract: Aligning robot policies with human preferences is essential for deployment to diverse end users. In per-user alignment approach, preference feedback is

ExToken: Structured Exploration for Efficient Vision-Language-Action Reinforcement Fine-tuning

arXiv:2607.12931v1 Announce Type: new Abstract: Reinforcement Learning (RL) has demonstrated significant potential for improving Vision-Language-Action (VLA) models on complex manipulation tasks. Howe

FlowWAM: Optical Flow as a Unified Action Representation for World Action Models

arXiv:2607.13017v1 Announce Type: cross Abstract: World Action Models (WAMs) are able to leverage pretrained video generators for both world modeling and action prediction. However, directly leveragin

From Geometric Recovery to Causal Validation: A Reproducible Audit of Sparse Autoencoder Features, from Superposition Geometry to Causal Inertness

arXiv:2607.12166v1 Announce Type: new Abstract: Sparse autoencoders (SAEs) are the standard for decomposing superposed neural representations into interpretable features, and evaluation relies predomi

Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models

arXiv:2607.12463v1 Announce Type: new Abstract: Coding agents must integrate external tool returns into ongoing reasoning - a capability that standard left-to-right pretraining on code exposes only in

GaitSpan: Growing Humanoid Locomotion from Walking to Running

arXiv:2607.12114v1 Announce Type: cross Abstract: A humanoid that can walk should not relearn locomotion from scratch to jog or run. Yet current approaches often obtain gait diversity by prescribing g

Model Releases

Generic AI models are not built for medicine. Doximity Ask, trained and served on Fireworks, outperformed GPT-5.6 Sol, Claude Fable 5, and O…

model-releasesfireworks-ai--x

Generic AI models are not built for medicine. Doximity Ask, trained and served on Fireworks, outperformed GPT-5.6 Sol, Claude Fable 5, and OpenEvidence in a Stanford-Harvard clinical AI safety study.

Gradient Flow Dynamics and Implicit Bias of Diagonal Linear Networks under Infinitesimal Initialization

arXiv:2607.12332v1 Announce Type: new Abstract: We study the gradient flow dynamics of diagonal linear networks for regression tasks under infinitesimal initialization. Extending Theorem 1 from Pesme

Gradient-free learning of a closed-loop wall controller for turbulent drag reduction

arXiv:2607.12626v1 Announce Type: cross Abstract: Closed-loop wall control learnt by multi-agent reinforcement learning can lower skin-friction drag in turbulent channels, but these gradient-based pol

Growing a Tail: Increasing Output Diversity in Large Language Models

arXiv:2411.02989v2 Announce Type: replace Abstract: How diverse are the outputs of large language models when diversity is desired? We examine the diversity of responses of several language models to

Hallo4D: Multi-Modal Hallucination Mitigation for Consistent Spatio-Temporal Generation

arXiv:2607.12752v1 Announce Type: cross Abstract: While recent advances in 3D generation have enabled impressive visual synthesis, existing methods often rely on 2D diffusion supervision without expli

Model Releases

How to Analyze and Govern Gemini Enterprise App Usage at Scale with BigQuery

model-releasesgoogle-cloud-ai

Deploying the Gemini Enterprise app across an organization marks a transformative leap forward in workforce productivity, providing employees with an amazing, high-performance suite of agentic AI tool

HSEmotion Team at the 11th ABAW Challenge: Multi-Task Learning and Ambivalence/Hesitancy Video Recognition

arXiv:2607.12774v1 Announce Type: cross Abstract: This article presents our results for the 11th Affective Behavior Analysis in-the-Wild (ABAW) competition. For multi-task learning with simultaneous p

IDC: Why the right networking approach is foundational to agentic AI

Editor’s note: Today we hear from IDC on the results of its 2026 AI in Networking Special Report Survey exploring the enterprises' concerns about networking infrastructure to support the rise of agent

safetygoogle-cloud-ai

I'm Sorry, but I Can't Help with Braille: Revealing Accessibility Failures in State-of-the-Art LLMs

arXiv:2607.11893v1 Announce Type: cross Abstract: Large Language Models (LLMs) perform strongly on many language tasks, but their capability in structurally constrained, accessibility-critical modalit

Imputation-free transformer learning enables robust Alzheimer's disease prediction and calibrated uncertainty quantification across heterogeneous clinical cohorts

arXiv:2607.11656v2 Announce Type: replace-cross Abstract: Accurate diagnostic classification and disease-severity prediction for Alzheimer's disease are hampered by the incompleteness and heterogeneit

JADE: Expert-Grounded Dynamic Evaluation for Open-Ended Professional Tasks

arXiv:2602.06486v4 Announce Type: replace Abstract: Evaluating agentic AI on open-ended professional tasks faces a fundamental dilemma between rigor and flexibility. Static rubrics provide rigorous, r

Learning When to Trust in Contextual Social Bandits

arXiv:2603.13356v2 Announce Type: replace Abstract: Robust reinforcement learning typically assumes that feedback sources are either globally trustworthy or corrupted within a fixed global budget. We

Lost in Visual Translation: A VLM-Assisted Perceptual-Semantic Coherence Framework for EEG-to-Image Reconstruction

arXiv:2607.12364v1 Announce Type: cross Abstract: EEG-to-image evaluation should distinguish visual fidelity from recoverable meaning. Yet EEG-derived reconstructions are blurry, distorted, and low-de

M2I2HA: Multi-modal Object Detection Based on Intra- and Inter-Modal Hypergraph Attention

arXiv:2601.14776v3 Announce Type: replace Abstract: Recent advances in multi-modal detection have significantly improved detection accuracy in challenging environments (e.g., low light, overexposure).

MAMMOTH: A Multi-Modal End-to-End Policy for Off-Road Mobility Robust to Missing Modality

arXiv:2607.12965v1 Announce Type: new Abstract: Reliable autonomous navigation in unstructured off-road environments remains a critical unsolved challenge due to extreme terrain diversity, drastic ill

MESH: Scaling Up Retrieval with Heterogeneous Content Unification

arXiv:2607.12392v1 Announce Type: cross Abstract: Optimizing large-scale retrieval hinges on the ability to efficiently surface candidates across diverse content tiers. However, to capture segments su

Mistake gating leads to energy and memory efficient continual learning

arXiv:2604.14336v2 Announce Type: replace Abstract: Synaptic plasticity is metabolically expensive, yet animals continuously update their internal models without exhausting energy reserves. However, w

MUSA-PINN: Multi-scale Weak-form Physics-Informed Neural Networks for Fluid Flow in Complex Geometries

arXiv:2603.08465v3 Announce Type: replace Abstract: While Physics-Informed Neural Networks (PINNs) offer a mesh-free approach to solving fluid-flow PDEs, standard point-wise residual minimization suff

“one 80% price cut ends OpenAI and Anthropic” possibly overstated but directionally correct.

“one 80% price cut ends OpenAI and Anthropic” possibly overstated but directionally correct. Everyone is debating the price of tokens. The bigger question is why they keep getting cheaper. As AI syste

safetygary-marcus--x

Operationalising Multi-Dimensional Evaluation for Conversational Agents: A Scalable, Governed Pipeline with Selective Re-evaluation and Model Benchmarking

arXiv:2607.12085v1 Announce Type: new Abstract: Evaluating retail conversational agents requires methods beyond lexical-overlap metrics to assess intent alignment, factuality, helpfulness, clarity, to

Policy-Conditioned Constrained Decoding for Column-Level Access Control in Text-to-SQL

arXiv:2607.12341v1 Announce Type: new Abstract: Text-to-SQL is increasingly deployed across trust boundaries between data providers and users. Such deployment must balance three competing requirements

PoseAlign: Sculpting Pose-Consistent Meshes via Text-Guided Deformation

arXiv:2607.10560v2 Announce Type: replace-cross Abstract: Mesh deformation, the process of altering the vertex positions of a 3D mesh while preserving its topological structure, is a cornerstone of co

Model Releases

Resist and Update: Counterfactual Report Coordinates for Incentive-Compatible LLMs

model-releasesarxiv-cs-ai

arXiv:2607.12985v1 Announce Type: new Abstract: Aligned language models routinely misreport under non-evidential incentive pressure: they agree with a confident user or overstate certainty even when t

Robot Drummer: Learning Rhythmic Skills for Humanoid Drumming

arXiv:2507.11498v3 Announce Type: replace Abstract: Humanoid robots have seen remarkable advances in dexterity, balance, and locomotion, yet their role in expressive domains such as music performance

Sat2RealCity: Geometry-Aware and Appearance-Controllable 3D Urban Generation from Satellite Imagery

arXiv:2511.11470v3 Announce Type: replace Abstract: 3D urban generation from satellite imagery is an important task for scalable digital twins and real-world simulation environments. Existing approach

Scalable Optimal Transport Algorithm for Network Alignment

arXiv:2607.11952v1 Announce Type: new Abstract: Network alignment identifies node correspondences across different networks and is a fundamental primitive in many data science applications, including

Structure-Semantic Co-optimized Latent Diffusion Model for Fast Visual Anagram Synthesis

arXiv:2606.16241v3 Announce Type: replace Abstract: Visual anagram is an intriguing form of art creation wherein a single image presents different conceptual interpretations under transformations such

TADPO: Reinforcement Learning Goes Off-road

arXiv:2603.05995v2 Announce Type: replace-cross Abstract: Off-road autonomous driving poses significant challenges such as navigating unmapped, variable terrain with uncertain and diverse dynamics. Ad

Text-Aided Multi-Modal Panoptic Symbol Spotting for CAD Floor Plan Drawings

arXiv:2607.12678v1 Announce Type: cross Abstract: Computer-Aided Design (CAD) floor plan drawings contain both graphical primitives and textual annotations, which provide complementary geometric and s

The creation of a regulatory body to enforce and implement standards, as suggested here by @demishassabis or by @AnthropicAI last month, wou…

The creation of a regulatory body to enforce and implement standards, as suggested here by @demishassabis or by @AnthropicAI last month, would be a huge step in the right direction to make AI developm

safetyyoshua-bengio--x