safety

Dynamic Lagging using Stable-Prefix Training for Simultaneous Translation

arXiv:2609.05799v2 Announce Type: replace Abstract: In streaming simultaneous speech translation, the speech translation system is trained to learn a read-write policy that alternates between consumin

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Efficient Bayes-Adaptive Reinforcement Learning with Temporal Logic Specifications

DGX agent

arXiv:2609.20954v1 Announce Type: new Abstract: We present a novel end-to-end model-based Reinforcement Learning (RL) algorithm for efficient policy synthesis under given Linear Temporal Logic (LTL) s

safetyarxiv-cs-lg

21 Sept 2026

ExpBoN: Exponential-Noise Best-of-n for Efficient Test-Time LLM Alignment

arXiv:2609.21899v1 Announce Type: new Abstract: Best-of-n (BoN) sampling is a simple yet effective inference-time alignment method, but hard maximization provides only coarse control over the trade-of

FedeRage: Provably Convergent Agnostic Federated Learning under General Client Drift

arXiv:2609.21057v1 Announce Type: new Abstract: Federated learning (FL) enables collaborative model training without sharing raw data, but its performance degrades under non-IID data and stochastic cl

Federated Deep Clustering Networks for High-Dimensional and Heterogeneous Data

arXiv:2609.21829v1 Announce Type: cross Abstract: Clustering high-dimensional data is a fundamental task in unsupervised machine learning with applications to a variety of domains. In the centralized

Fewer Steps, Better Actions: Rethinking Flow-Matching Inference for VLA Policies

arXiv:2609.21216v1 Announce Type: cross Abstract: Vision-language-action (VLA) policies based on flow matching generate action chunks through repeated evaluations of an action expert. Increasing the n

FootQuery: Future-Touchdown-Guided Retrieval from Depth History for Perceptive Humanoid Locomotion

arXiv:2609.21447v1 Announce Type: cross Abstract: Humanoid locomotion over complex terrain requires anticipating footholds that may no longer be visible at touchdown. Limited camera coverage and self-

From Pretraining to Proficiency: Real-World Subtask RL for Long-Horizon Manipulation with Minimal Human Intervention

arXiv:2609.21788v1 Announce Type: cross Abstract: A pretrained robot foundation policy may execute most of a long-horizon task yet repeatedly fail at a few critical subtasks. Collecting additional ful

GEM-MPC: Balancing Exploration and Exploitation through Expert-Guided Planning

arXiv:2609.21735v1 Announce Type: new Abstract: Effective exploration in high-dimensional continuous control remains a central challenge in reinforcement learning. Planning-based methods address this

Generative Artificial Intelligence Chatbots for Motivational Interviewing: A Scoping Review From System Design to Intervention Outcomes

arXiv:2609.20902v1 Announce Type: new Abstract: Motivational interviewing (MI) is a collaborative approach to elicit autonomous motivation for health behavior change. Generative AI (GenAI) offers new

Geometric Mean Pooling for Equal-Weight Multiplicative Coarse-Graining

arXiv:2609.21876v1 Announce Type: new Abstract: As an alternative to the additive and extremal biases of average and max pooling, we introduce Geometric Mean Pooling (GMP), a signed pooling operator t

Goal-Oriented Reactive Simulation for Closed-Loop Trajectory Prediction

arXiv:2603.24155v3 Announce Type: replace Abstract: Current trajectory prediction models are primarily trained in an open-loop manner, which often leads to covariate shift and compounding errors when

Gradient-Stable Attention Heads Signal LLM Correctness

arXiv:2602.13699v2 Announce Type: replace Abstract: Large language models (LLMs) often generate plausible yet incorrect answers, posing risks in safety-critical settings such as medicine. Human evalua

GVPO++: Group Variance Policy Optimization for LLM Post-Training and On-Policy Distillation

arXiv:2609.21432v1 Announce Type: new Abstract: Post-training plays a pivotal role in enhancing the reasoning capabilities and task-specific expertise of large language models (LLMs). Despite recent a

HAT: Hypothesis-Anchored Tracking for Video Monocular Spacecraft Pose Estimation

arXiv:2609.21597v1 Announce Type: new Abstract: Monocular 6-DoF pose estimation of non-cooperative targets is important for on-orbit servicing and debris removal. A single-image estimator can confuse

Hiding in Plain Sight: A Diffusion-based Mitigation of Geolocation Privacy Leakage in Vision-Language Models

arXiv:2609.21363v1 Announce Type: new Abstract: Multimodal large reasoning models (MLRMs) have demonstrated remarkable capabilities in complex visual understanding. However, this very power introduces

IncentRL: The Trade-Off Between Preference Guidance and Task Performance

arXiv:2609.21525v1 Announce Type: new Abstract: Preference-based reward shaping can guide reinforcement learning, but adding preference signals to the reward may unintentionally change the task being

Info3R: Information-Adaptive Test-Time Training for 3D Reconstruction

arXiv:2609.21938v1 Announce Type: new Abstract: Transformer-based models have recently achieved strong performance on 3D reconstruction from images, and recent works extend them to process video strea

Information-Gain Rewards over Diversity-Pruned Tests: GT-Anchored Verifier Co-Training for Reliable Code Generation

arXiv:2609.21208v1 Announce Type: new Abstract: Self-play methods that co-train a single language model as both coder and test author promise to move code-generation RL beyond fixed test suites, but t

K2-V2: A 360-Open, Reasoning-Enhanced LLM

arXiv:2512.06201v3 Announce Type: replace Abstract: We introduce K2-V2, a 360-open LLM built from scratch as a superior base for reasoning adaptation, in addition to functions such as conversation and

KnowDemo: Knowledge-Guided Robot Demonstration Generation from Human Videos

arXiv:2609.21229v1 Announce Type: cross Abstract: Learning robot manipulation policies typically requires substantial demonstration data, which are costly to collect on real robots. Recent methods gen

lambda-Controlled GRPO: Turning Flow-Matching Ratio Instability into a Budgeted Resource

arXiv:2609.22041v1 Announce Type: new Abstract: Reinforcement learning is increasingly used to align image generators with reward signals, and Flow-GRPO recently extended this paradigm to flow-matchin

Learning Distance-Conditioned Object Transport for Humanoid Loco-Manipulation from a Single Motion Clip

arXiv:2609.21467v1 Announce Type: new Abstract: Motion tracking can reproduce humanoid loco-manipulation from a single retargeted motion clip, but a policy trained on a fixed reference primarily repro

Learning End-to-End Control for Omnidirectional Aerial Motion on Overactuated Tilt-rotor Quadrotors

arXiv:2602.21583v2 Announce Type: replace Abstract: While reinforcement learning (RL) has been successfully applied to conventional quadrotors for agile and robust flight, whether actuator-level RL ca

Learning Scene-Aware Humanoid Locomotion through 3D Clutter from Immersive Human Demonstrations

arXiv:2609.21107v1 Announce Type: new Abstract: While learning from human motions has enabled highly dynamic humanoid skills such as dancing and martial arts in obstacle-free space, traversal through

Let’s be clear: LLMs don’t feel pain. And they don’t deserve moral standing. This new paper reports a series of experiments identifying pain…

Let’s be clear: LLMs don’t feel pain. And they don’t deserve moral standing. This new paper reports a series of experiments identifying pain-related representations in LLMs. The authors also show that

safetygary-marcus--x

LIMBO: Learning and Internalizing Model-Free Barrier Objectives for Agile and Safe Whole-Body Control

arXiv:2609.22075v1 Announce Type: new Abstract: Safe whole-body control requires coordinating collision avoidance and balance under high-dimensional, nonlinear dynamics--making safety certificates dif

M2G-LLM: Enhancing Clinical Prediction via Multimodal Graph Reasoning and LLM Context Injection

arXiv:2609.21164v1 Announce Type: new Abstract: Integrating diverse data modalities --- such as clinical notes, laboratory results, and medical imaging --- is essential for advancing clinical decision

MetaPusher: Meta Learning and Planning for Nonprehensile Manipulation of Unseen Objects with Rapid Online Adaption

arXiv:2609.21122v1 Announce Type: new Abstract: Manipulating previously unseen objects remains challenging, as their dynamics depend on latent physical properties, such as friction and mass distributi

MicroHookACT: Monocular Microscopic Vision Guided Visuomotor Policy for Flexible Microelectrode Hooking

arXiv:2609.21365v1 Announce Type: new Abstract: Automated needle-loop hooking is a critical step in flexible microelectrode (FME) implantation. This paper presents MicroHookACT, an imitation learning-

MintAct: A Unified Visual Agent for Digital Environments

arXiv:2609.22083v1 Announce Type: new Abstract: We present MintAct, a family of vision-language models that unifies UI grounding, multi-step navigation across mobile, desktop, and web, and visual tool

MIST: Multimodal Survival Prediction with Genomic-Guided Histology Attention

arXiv:2609.21811v1 Announce Type: new Abstract: Multimodal survival models can combine complementary prognostic information from whole-slide images and genomic profiles, but effective fusion remains c

Moral Entropy: Auditing Bias and Uncertainty in Moral Judgment

arXiv:2609.21992v1 Announce Type: new Abstract: Most work in computational ethics treats annotator disagreement on moral content as noise to be voted away, collapsed into majority vote or the more per

MT-WAM: Reorienting the One-Pass Predictive Representation Toward Action Generation

arXiv:2609.21474v1 Announce Type: new Abstract: Fast-WAM shows that video-action co-training improves control without generating future video at inference, making the representation from a single vide

NaViRrator: Robot Navigation from Human-Readable Maps through a Learned Visual Route

arXiv:2609.21316v1 Announce Type: new Abstract: Human-readable maps provide an intuitive interface for specifying robot destinations, but connecting their schematic geometry to egocentric observations

Offline Constrained RLHF with Multiple Preference Oracles

arXiv:2604.00200v2 Announce Type: replace Abstract: We study offline constrained reinforcement learning from human feedback with multiple preference oracles. Motivated by applications that trade off p

On Repulsive and Attractive Teachers: Separating Correctness from Behavior in Self-Distillation

arXiv:2609.21561v1 Announce Type: cross Abstract: On-policy self-distillation provides dense, token-level supervision by conditioning a model on privileged information and distilling the resulting tea

On the Limits of Maximal Coding Rate Reduction for Out-of-Distribution Generalisation

arXiv:2609.21001v1 Announce Type: new Abstract: Substantial efforts have been devoted to making deep learning objectives, representations, and architectures interpretable, with the goal of improving t

OneBid: A Unified Auto-Bidding Foundation Model for Diverse oCPX Advertising Scenarios

arXiv:2609.21550v1 Announce Type: cross Abstract: Auto-bidding is central to computational advertising, where strategies must maximize advertisers' conversion value under economic constraints. It has

Outcome-Conditioned End-Effector Geometry Across Vision-Language-Action Policies

arXiv:2609.21659v1 Announce Type: cross Abstract: Vision-language-action (VLA) policies solve the same manipulation task through different action interfaces, but task success alone does not establish

OverThink: Slowdown Attacks on Reasoning LLMs

arXiv:2502.02542v5 Announce Type: replace Abstract: A reasoning language model (RLM) generates costly reasoning tokens, often hidden from the users, that help it excel at many tasks. Our Overthink att

Potential-Field Action Representation for Reinforcement Learning in Contact-Rich Manipulation

arXiv:2609.21609v1 Announce Type: cross Abstract: Model-free reinforcement learning can acquire contact-rich robotic manipulation skills through trial-and-error interaction, but it often requires the

Predictable Failure in Multi-Hop Retrieval: Score-Distributional Confidence Scoring and Abstention

arXiv:2609.22056v1 Announce Type: cross Abstract: Multi-hop retrieval failures are not uniformly distributed across queries: they cluster in structurally predictable subpopulations. We prove two resul

PrismAlign: Prior-Steered Multi-View VLM Alignment for Hallucination-Robust Table OCR

arXiv:2609.21351v1 Announce Type: new Abstract: Table extraction suffers from frequent structural errors and semantic hallucinations. We propose PrismAlign, a multi-VLM framework aligning diverse visu

Provably Optimal Reinforcement Learning under Safety Filtering

arXiv:2510.18082v3 Announce Type: replace Abstract: Recent advances in reinforcement learning (RL) enable its use on increasingly complex tasks, but the lack of formal safety guarantees still limits i

PSR: Predictive Sensorimotor Representation Learning for Contact-Rich Manipulation

arXiv:2609.21753v1 Announce Type: new Abstract: Contact-rich manipulation requires policies to generate precise actions by reasoning over contact forces, robot configurations, and interaction historie

RAYA: Learning Where and When to Intervene for Robot Recovery

arXiv:2609.21690v1 Announce Type: new Abstract: A robot can predict failure and still be unable to prevent it. By the time a safety mechanism reacts, the nominal plan may already have spent the contro

Rectify, Don't Regret: On-Policy Closed-Loop Training for Multimodal Trajectory Prediction

arXiv:2603.23393v2 Announce Type: replace Abstract: Current trajectory prediction models are primarily trained in an open-loop manner, which often leads to covariate shift and compounding errors when