safety

From Rebound to Remedy: Understanding and Mitigating Reward Hacking via Representation Engineering

arXiv:2604.01476v2 Announce Type: replace-cross Abstract: Reinforcement learning for LLMs is vulnerable to reward hacking, where models exploit shortcuts to maximize reward without solving the intende

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

From Semantic Grounding to Decision Optimization: A Unified Framework for Long-Horizon UAV Vision-Language Navigation

DGX agent

arXiv:2608.09564v1 Announce Type: cross Abstract: UAV vision-language navigation (UAV-VLN) focuses on enabling an aerial agent to follow natural-language instructions in open 3D environments from egoc

safetyarxiv-cs-ai

11 Aug 2026

From @sequoia's own your intelligence event. We're starting to see a new category of company emerge, the Full-Stack AI company, where innova…

From @sequoia's own your intelligence event. We're starting to see a new category of company emerge, the Full-Stack AI company, where innovation happens at both the product and intelligence layer. Tha

safetysonya-huang--x

From Single Chatbots to Governed Agent Ecosystems: An Agentic AI Pattern Catalogue and Orchestration Framework for Mission-Critical Hospital Information Management Systems

arXiv:2608.07627v1 Announce Type: new Abstract: Hospitals are racing to embed AI, while coping with the surge in adaptation of the technology in other industries, into the triage management, documenta

Generalized Convexity and Smoothness via Conjugate Duality: Optimization Theory for Deep Neural Networks

arXiv:2608.09523v1 Announce Type: new Abstract: Deep neural network (DNN) training with stochastic gradient descent (SGD) and its variants achieves strong empirical performance, yet classical optimiza

Generalizing deep reinforcement learning across cable-driven parallel robot configurations with actuator-level policies

arXiv:2608.07546v1 Announce Type: cross Abstract: Cable-driven parallel robots (CDPRs) present diverse configurations and complex control challenges, which can be addressed by deep reinforcement learn

GRACE: LLM-Grounded Semantic Metric Spaces for Scalable Mixed-Data Clustering

arXiv:2608.07881v1 Announce Type: new Abstract: Clustering mixed tabular data requires a unified metric space to bridge the inherent heterogeneity between continuous numerical measurements and discret

Graph-Guided Safe Diffuser: Topological Graph Guidance for Safe Diffusion Planning

arXiv:2608.09484v1 Announce Type: new Abstract: Many diffusion-based planners enforce safety through inference-time guidance, but such interleaved trajectory deformations often degrade kinematic feasi

Hierarchical Topology-Aware Planning and Control of Underwater Vehicle-Manipulator Systems in Confined Environments

arXiv:2608.08871v1 Announce Type: new Abstract: This paper addresses autonomous intervention with an underwater vehicle--manipulator system (UVMS) in confined, cluttered, and partially known environme

High-Capacity Generalized Hopfield Networks

arXiv:2608.08226v1 Announce Type: cross Abstract: Generalized Hopfield networks are introduced where memories and neurons are continuous variables that lie on a Riemannian manifold. We explicitly focu

HoloAegis: Frozen Representation, Topological Inference: Minimally Parametric Safety Manifolds for Zero-Shot LLM Guardrails

arXiv:2608.08485v1 Announce Type: new Abstract: Current LLM safety guardrails face a fundamental tension: fine-tuning distorts pre-trained representations while generative judges incur prohibitive inf

HonestFace: Towards Honest Face Restoration with One-Step Diffusion Model

arXiv:2505.18469v2 Announce Type: replace Abstract: Face restoration has achieved significant advancements through the years of development. However, maintaining high fidelity and authenticity while a

Illusion of Alignment: Detecting Hidden Disagreement in Collaborative Dialogue

arXiv:2608.08210v1 Announce Type: new Abstract: Collaborative dialogue can end with apparent agreement while participants still differ on goals, assumptions, or execution plans, creating an extbf{illu

Impact of Dataset Composition on Embedded Real-Time UAV Wildfire Detection Using Compact YOLO Models

arXiv:2608.07554v1 Announce Type: new Abstract: The development of vision-based wildfire detection systems for unmanned aerial vehicles is constrained by the limited availability of diverse real-world

Improving Generalization Robustness of Multimodal RLVR

arXiv:2608.08802v1 Announce Type: new Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) makes Multimodal Large Language Models more accurate, but the gains are brittle: simply paraphrasi

In-Loop Model Adaptation with Coupled Latent-Noise Guidance for High-Fidelity Subject-Driven Text-to-Image Generation

arXiv:2608.09244v1 Announce Type: new Abstract: Text-to-image diffusion models have achieved remarkable success in generating high-quality images from a given text prompt. Subject-driven generation ai

Integrating RCTs, RWD, AI/ML and Statistics: Next-Generation Evidence Synthesis

arXiv:2511.19735v2 Announce Type: replace-cross Abstract: Randomized controlled trials (RCTs)have been the cornerstone of clinical evidence; however, their cost, duration, and restrictive eligibility

Intent Speaks Louder: Controllable User Simulation Beyond Response Imitation

arXiv:2608.09420v1 Announce Type: new Abstract: User simulators are widely used as scalable environments for training and evaluating interactive assistants. Generating the next user turn is inherently

Interpreting Video Representations with Spatio-Temporal Sparse Autoencoders

arXiv:2604.03919v2 Announce Type: replace-cross Abstract: We present the first systematic study of Sparse Autoencoders (SAEs) on video representations. Standard SAEs decompose video into interpretable

Investigating Multimodal Informativity under Different Partner Visibility Conditions in Video-Mediated Dialogue

arXiv:2608.08915v1 Announce Type: new Abstract: Situated language use is multimodal and embodied. For example, gestures can carry information that is absent or underspecified in the speech signal, yet

Is CLIP Cross-Eyed? Revealing and Mitigating Center Bias in the CLIP Family

arXiv:2604.05971v2 Announce Type: replace-cross Abstract: Recent research has shown that contrastive vision-language models such as CLIP often lack fine-grained understanding of visual content. While

JEPA-WAM: Learning Vision-Language-Action Policies with Joint-Embedding World Modeling

arXiv:2608.09381v1 Announce Type: new Abstract: Robust robot control benefits from explicitly modeling state transitions, but video-generation world action models (WAMs) introduce substantial deployme

JustLLMGRPO: Radiographic Control for Chest X-Ray Generation

arXiv:2608.08046v1 Announce Type: new Abstract: Text-conditioned chest X-ray generation aims to synthesize realistic radiographs that faithfully depict specified findings. Existing work has primarily

KGCache: Amortized Subgraph Retrieval for KG Reasoning with LLMs

arXiv:2608.07954v1 Announce Type: new Abstract: Large language models can answer knowledge-intensive questions more reliably when they are grounded with knowledge graphs, but systems such as Think-on-

Knowing When to Ask: Resolving Uncertainty in Human-Robot Joint Planning via Explicit Dialogue and Implicit Intent Cues

arXiv:2603.07822v2 Announce Type: replace Abstract: Effective human-robot collaboration in open-world environments requires joint planning under uncertainty about the task, the environment, and the hu

Knowledge-Distilled End-to-End Reinforcement Learning for Smooth 6-DOF Thrust Control and Rapid Adaptation to Ocean Currents in Remotely Operated Vehicles

arXiv:2608.08598v1 Announce Type: new Abstract: With the continuous improvement of computational capabilities, end-to-end reinforcement learning has been rapidly developed for remotely operated vehicl

KumbhDoot: A Scale-Ready, LLM-Bounded Architecture for Mass-Gathering Public-Service Assistants

arXiv:2608.07520v1 Announce Type: cross Abstract: Mass religious gatherings such as the Kumbh Mela concentrate tens of millions of people into a single region over a few weeks, producing intense, repe

LASA: Language-and-Source-Anchored Alignment for Domain Generalized Semantic Segmentation

arXiv:2608.08805v1 Announce Type: new Abstract: Domain Generalization Semantic Segmentation (DGSS) focuses on generalizing knowledge from labeled source domains to unseen target domains where data is

Learning aligned EEG representations with subject-specific encoders

arXiv:2606.16462v2 Announce Type: replace-cross Abstract: Cross-subject EEG decoding promises more training data, but it also exposes neural networks to strong inter-subject distribution shifts. We st

Learning an Interior Layout Policy in a Domain Specific Language Action Space

arXiv:2608.07547v1 Announce Type: cross Abstract: Indoor scene layout generation is a challenging task in interior design. Existing methods often oversimplify the task by reducing room conditions to c

Learning Deep Modality-Shared Self-Expressiveness for Image Clustering with Textual Information

arXiv:2608.08418v1 Announce Type: new Abstract: Leveraging textual information for image clustering has emerged as a promising direction, largely owing to the powerful representations learned by Visio

Learning from Consensus and Disagreement: Unsupervised On-Policy Self-Distillation with Minority-Trajectory Contrast

arXiv:2608.08764v1 Announce Type: cross Abstract: On-policy self-distillation improves language-model reasoning by querying a teacher on states actually visited by the student. Recent methods create a

Learning Multi-Timescale Interventions under Safety and Resource Constraints

arXiv:2508.03875v2 Announce Type: replace Abstract: Many sequential decision problems offer qualitatively different ways of influencing the environment: some interventions act immediately, whereas oth

Learning Preference Adaptation for Large Language Model Personalization via Verbal Reinforcement Learning

arXiv:2608.09507v1 Announce Type: cross Abstract: Natural language user preferences provide an interpretable interface for LLM personalization. However, universal preference summaries often contain in

Learning to Coordinate Symbolic Tools: LLM Agents for Verified Sum-of-Squares Certificates

arXiv:2608.00326v2 Announce Type: replace Abstract: Tool calling allows large language models (LLMs) to invoke external computation during problem solving, a useful capability in various fields includ

Learning to Modulate, Not to Cycle: Soft Actor---Critic Recovers Inverter-Style Heat-Pump Control

arXiv:2608.09453v1 Announce Type: cross Abstract: On--off cycling is the main cause of compressor wear in residential heat pumps, yet reinforcement learning (RL) controllers for buildings typically op

Learning When to See and When to Feel: Adaptive Vision-Torque Fusion for Contact-Aware Manipulation

arXiv:2604.01414v2 Announce Type: replace Abstract: Vision-based policies have achieved a good performance in robotic manipulation due to the accessibility and richness of visual observations. However

Legal Responsibilities Using Autonomous Agents For Artificial Intelligence

arXiv:2608.08022v1 Announce Type: new Abstract: Recent incidents involving Artificial Intelligence (AI) agents, which were reported escaping their containment `unintentionally' to gain unauthorized ac

LF{}^{2}AR: Accounting for Layerwise Dynamics to Improve Multimodal Adaptation of Language Models

arXiv:2503.06211v3 Announce Type: replace-cross Abstract: Text-pretrained language models (LMs) encode rich world knowledge, but adapting them to process and generate perceptual modalities such as aud

Linguistically-Aligned and Visually-Grounded Preference Optimization for Clinically-Augmented Medical Report Generation

arXiv:2608.08494v1 Announce Type: new Abstract: Despite significant advances in Medical Report Generation (MRG), the reliability remains constrained by the prevalence of factual errors. While Direct P

LLM Reasoning for Subjective Tasks: Failure Modes, Mitigation, and Dynamic Reasoning Routing

arXiv:2608.08889v1 Announce Type: new Abstract: Recommendation systems thrive on personalization, where ''correctness'' is rarely a binary truth but a matter of subjective human preference. As Large L

LLMs Remember First, Forget Last: Dual-Process Interference in Large Language Models

arXiv:2603.00270v3 Announce Type: replace-cross Abstract: Large language models can process millions of tokens, yet how they handle conflicting information within context remains poorly understood. Fr

LUCID: Latent-Skill Unified Control via Imagined Dynamics for Long-Horizon Humanoid Loco-Manipulation

arXiv:2608.07746v1 Announce Type: new Abstract: Long-horizon humanoid loco-manipulation requires composing versatile whole-body skills and reliable high-level decision making. Existing methods often c

Machine Learning and Data Analysis Using Posets: A Survey

arXiv:2404.03082v3 Announce Type: replace Abstract: Partially ordered sets (posets) are discrete mathematical structures that formalize the notion of comparison without forcing every pair of objects t

Machine-Learning-Based Diagnostic Framework for Passive Ultrasonic Detection of Railway Wheel Defects

arXiv:2608.08301v1 Announce Type: new Abstract: Reliable identification of railway wheel defects is important for safety and maintenance. This study develops a machine-learning-based diagnostic framew

MAGIC-SSCIL: Manifold Anchoring and Geometric Incremental Calibration for Semi-Supervised Class Incremental Learning

arXiv:2608.07586v1 Announce Type: new Abstract: Semi-supervised Class Incremental Learning (SSCIL) is a severe challenge for neural networks, and it is hardest in the exemplar-free setting where no pa

MARA: Flow-Matching-Guided Multi-Agent Resource Allocation for Computational Resource Efficient Learning

arXiv:2608.09130v1 Announce Type: cross Abstract: Allocating limited computation among concurrent learning tasks is difficult when each task must reach a target loss before a deadline but its required

Matching Supervision to the Student's Learning Capacity: A Unified Framework for On-Policy Self-Distillation

arXiv:2608.08176v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) improves the reasoning abilities of LLMs by internalizing privileged context into model parameters through self-disti