safety

Game-Guided Skill Discovery through Self-Play for Playable Agent Control

arXiv:2609.40137v1 Announce Type: cross Abstract: We present Game-Guided Skill Discovery (GGSD), a framework that uses self-play in games to discover motor skills that are directly playable by humans.

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

Gary Marcus @GaryMarcus (Emeritus Professor, New York University) took the stage at TAIS 2026 and made the case that, 'LLMs are not the way …

DGX agent

Gary Marcus @GaryMarcus (Emeritus Professor, New York University) took the stage at TAIS 2026 and made the case that, 'LLMs are not the way to alignment.' Gary argued that LLMs are constitutionally il

safetygary-marcus--x

1 Oct 2026

GFD-OPD: Guidance-Folded On-Policy Distillation of Diffusion Models Across Scales

arXiv:2609.39692v1 Announce Type: cross Abstract: On-policy distillation (OPD) has demonstrated two important capabilities in language models: compressing large teachers into smaller students and merg

GLARE: Generating Listening Heads with Appropriate Reactions

arXiv:2609.40317v1 Announce Type: new Abstract: While talking head generation has advanced rapidly, generating natural listener behavior in dyadic conversations, which know when to react, how to react

GraphCert: Bootstrap Agentic Graph Reasoning with Certified Evidence Rubrics

arXiv:2609.38798v1 Announce Type: new Abstract: Graph agents extend large language models (LLMs) with the ability to actively explore and reason over knowledge graphs through multi-step interactions w

Great paper from Meta on agent harness optimization. Meta-Harness-style search uses one development set and one proposal policy, so every ed…

Great paper from Meta on agent harness optimization. Meta-Harness-style search uses one development set and one proposal policy, so every edit follows a single path and can get stuck in a local optimu

safetydair-ai--x

Grokking through the Lens of Minimum-Norm Interpolation

arXiv:2609.38453v1 Announce Type: new Abstract: Grokking shows that fitting the training data and learning the underlying signal can occur at very different stages. However, existing theories offer li

HARDE: Optimizing Agent Harnesses for Runtime Risk Detection and Execution Control

arXiv:2609.38291v1 Announce Type: cross Abstract: Large language model (LLM) agents are vulnerable to safety risks such as injected malicious instructions or misleading information, motivating runtime

Here the World in Stereo: Learning Dynamic Spatial Correspondence for Immersive Joint Video-Audio Generation

arXiv:2609.38748v1 Announce Type: new Abstract: Recent joint video-audio generation models have achieved strong semantic correspondence and temporal synchronization. However, applications such as AR/V

I have never seen a company more on-brand than OpenAI, one unethical move after another.

I have never seen a company more on-brand than OpenAI, one unethical move after another. It's such a full circle moment for OpenAI to use their own AI agents, built on stolen data, to find new and une

safetygary-marcus--x

Improving Fairness of Large Language Model-Based ICU Mortality Prediction via Case-Based Prompting

arXiv:2512.19735v4 Announce Type: replace Abstract: Accurately predicting mortality risk in intensive care unit (ICU) patients is critical for clinical decision-making. Large language models (LLMs) ar

Improving OCR Faithfulness via Gated and Attenuated On-Policy Distillation

arXiv:2609.38282v1 Announce Type: new Abstract: Vision-language models may rewrite anomalous text in images into linguistically plausible expressions, compromising OCR transcription faithfulness. Sequ

In-Distribution Imagination for Model-Based Offline Reinforcement Learning

arXiv:2609.38673v1 Announce Type: new Abstract: Model-based offline reinforcement learning (MBORL) improves sample efficiency through model-generated trajectories. However, accumulative model error ca

Interactor: Agentic RL oriented Iterative Creation for Ad Description Generation in Sponsored Search

arXiv:2606.15911v2 Announce Type: replace Abstract: This paper focuses on automatically generating informative ad descriptions in sponsored search. Unlike ad titles which are usually optimized to attr

Is Better Teacher Supervision Enough? Unlocking Student-side Learning in Multimodal On-Policy Distillation

arXiv:2609.39120v1 Announce Type: new Abstract: On-policy distillation (OPD) improves reasoning by providing token-level supervision from a teacher on a student's own trajectories. Existing methods pr

It Takes Little to Rewrite Perception: Targeted Semantic Substitution in Vision-Language Models at epsilon leq 4/255

arXiv:2609.38298v1 Announce Type: new Abstract: Vision Language Models (VLMs) are widely deployed in safety-critical scenarios, and understanding to which extent they can be controlled by adversarial

ITO: Multi-View Alignment and Training-Time Fusion for Image-Text Pretraining

arXiv:2603.02767v4 Announce Type: replace-cross Abstract: Image--text contrastive pretraining has become a dominant paradigm for visual representation learning, yet existing methods often yield repres

It's incredibly important to aggressively resist the idea that AI can be conscious or have 'personhood,' not only because it's as ridiculous…

It's incredibly important to aggressively resist the idea that AI can be conscious or have 'personhood,' not only because it's as ridiculous as believing that a book can think, but because once this i

safetygary-marcus--x

LatentHarness: Learning Latent Actions for Memory and Reasoning via Counterfactual Policy Distillation

arXiv:2609.39740v1 Announce Type: new Abstract: Long-context reasoning faces two complementary bottlenecks: retaining evidence across long inputs and sustaining computation across many reasoning steps

LBDU-VIO: Learned Bias Dynamics and Uncertainty for Visual-Inertial Odometry with Unreliable Vision

arXiv:2609.39125v1 Announce Type: new Abstract: Visual-inertial odometry (VIO) for aerial robots relies on high rate inertial measurement unit (IMU) propagation between visual updates. However, conven

LEARN-TS: LLM-Enhanced Alignment and Reconstruction with Normality Guidance for Multivariate Time-Series Anomaly Detection

arXiv:2609.38789v1 Announce Type: new Abstract: Reconstruction errors in multivariate time-series anomaly detection may not reliably distinguish abnormal behavior from benign deviations. Language-deri

Learning Adaptive and Visually Aligned Conditions for Generative Zero-Shot Learning

arXiv:2603.06281v3 Announce Type: replace Abstract: Generative zero-shot learning (ZSL) synthesizes visual features for unseen classes by learning a semantic-conditioned generator from seen classes. S

Learning-Enabled Estimation: Tight Characterizations under Sample Selection Biases

arXiv:2609.38608v1 Announce Type: new Abstract: When can we learn from biased samples? We study regression when outcomes are observed only after passing through selection filters that depend on both c

Learning to Explain While Planning: Rule-Aligned Diffusion Planning for Autonomous Driving

arXiv:2609.39995v1 Announce Type: new Abstract: Diffusion planners exhibit strong capabilities in generating multimodal trajectories. However, existing methods primarily rely on expert demonstrations

Learning to Reason with Compressed Context: Ground-Truth-Free Adaptation of OmniLLMs via Self-Distillation

arXiv:2609.39953v1 Announce Type: new Abstract: Omni-modal large language models (OmniLLMs) enable unified audio-video understanding, but their long multimodal token sequences make deployment computat

Less Data, Better Timing: Student-Curriculum Coupling for VLM On-Policy Distillation in Temporal Video Grounding

arXiv:2609.40055v1 Announce Type: cross Abstract: On-policy distillation (OPD) provides dense supervision directly on student-generated trajectories, making it an effective post-training strategy for

LexReward: A Taxonomy-Driven Reward Framework for Legal Language Models

arXiv:2609.39071v1 Announce Type: new Abstract: Legal language models require reward signals that capture not only answer correctness but also the multidimensional quality of legal responses. Existing

Linear Recurrent Memory Suffices to Distil a World-Model Policy for Robot Air Hockey

arXiv:2609.39151v1 Announce Type: cross Abstract: Does memory-dependent control need nonlinear recurrent dynamics? We study simulated air-hockey defence under temporary loss of puck tracking. A Dreame

Local-Minimum Escaper: Programmatic Subgoal Generation for Robust Navigation in Unknown Environments

arXiv:2609.38928v1 Announce Type: new Abstract: Mapless navigation in unknown and partially observable environments remains challenging for mobile robots, particularly when local minima prevent the ro

Locomotion-Grounded Humanoid Soccer: Task-Gated Reinforcement Learning of a Multi-Directional Kicking Library

arXiv:2609.38852v1 Announce Type: new Abstract: Recent humanoid soccer systems make motion tracking the substrate and derive locomotion from it, typically by steering a motion-reference anchor toward

LocoWM: High-Precision Locomotion through World-Model-Guided Residual Adaptation

arXiv:2609.39179v1 Announce Type: new Abstract: High-precision locomotion combines motion-command tracking with precise regulation of task-relevant physical states, enabling robots to interact reliabl

Loop-Free Inverse Reinforcement Learning via Sequential Value Recovery with Q-Score Matching

arXiv:2609.38955v1 Announce Type: cross Abstract: Inverse Reinforcement Learning (IRL) aims to recover a reward function that explains expert demonstrations. Existing IRL methods typically rely on a b

Making Grid Beam Search Less Greedy

arXiv:2609.39368v1 Announce Type: new Abstract: A common formalism for constraining the output of autoregressive text generation models involves lexical constraints, words or phrases which are require

Making LLMs Say What They Think: Measuring and Improving CoT-Interpretability Alignment

arXiv:2609.38972v1 Announce Type: cross Abstract: Chain-of-thought (CoT) traces often serve as a proxy for how Large Language Models (LLMs) arrive at their answers. However, growing evidence shows tha

Making Waves: A Membrane-Coupled Delta Array for Manipulating Objects Below the Actuator Spacing

arXiv:2609.39652v1 Announce Type: new Abstract: Distributed manipulator systems manipulate objects through the coordinated motion of many actuators. However, an object must be supported by several act

MC-PanDA++: Simpler, Stronger, and More Robust Domain-Adaptive Panoptic Segmentation

arXiv:2609.39681v1 Announce Type: new Abstract: Unsupervised domain adaptation (UDA) reduces the annotation burden in panoptic segmentation by leveraging a cost-effectively labeled source domain (e.g.

MCD: Causal Distillation of Multimodal In-Context Learning in Large Vision-Language Models

arXiv:2609.39920v1 Announce Type: cross Abstract: Large vision-language models (LVLMs) exhibit strong multimodal in-context learning (ICL) capabilities, yet this ability degrades substantially as mode

Memorize, Adapt, Ignore: Diagnosing Robot Learning Mechanisms under Training Data Variation

arXiv:2609.38401v1 Announce Type: new Abstract: Training data variation, whether through designing a domain randomization (DR) scheme in simulation or curating demonstrations for imitation learning, i

MindWorldBench: Evaluating Mental-State-to-Behavior Reasoning in Image-to-Video Generation

arXiv:2609.39147v1 Announce Type: new Abstract: Current image-to-video models achieve visual realism and physical plausibility, but reasoning about mental states remains unexplored. Actions are driven

Mitigating the Length-Scaling Tax with Online Distillation

arXiv:2609.38854v1 Announce Type: cross Abstract: Length scaling during reinforcement-learning (RL) post-training is often viewed as a sign of improved reasoning ability, especially on difficult probl

More Choices, Fewer Decisions: Ordinal-Scale Bias in JEV-like Direct-Decision Models

arXiv:2609.38827v1 Announce Type: new Abstract: Direct-decision models turn text into low-latency structured labels and scores, making them attractive for classification and automatic evaluation. Yet

Multi-Link Safety Filtering for VLA Policies Around Moving Hazards

arXiv:2609.40007v1 Announce Type: cross Abstract: A vision-language-action (VLA) policy can finish a manipulation task while knocking over objects unrelated to it, so task success alone does not show

Multi-LLM Collaborative Alignment via Stackelberg Games

arXiv:2609.39076v1 Announce Type: new Abstract: A pool of language models can collaborate and improve collectively by learning from one another's responses. These interactions depend on the instructio

MVP-SLAM: Multi-Camera Visual-Inertial Floorplan-Prior SLAM

arXiv:2609.39596v1 Announce Type: new Abstract: Indoor building construction sites are demanding environments for visual SLAM, where variable lighting and repetitive, low-textured structures make the

no way will this government properly police companies that it has a 10% equity stake in

no way will this government properly police companies that it has a 10% equity stake in 🚨 BREAKING: Trump REJECTS nationalizing OpenAI and Anthropic but says the US government might take an Intel-styl

safetygary-marcus--x

OGPO: One-Step Generative Policy Optimization for Real-Time Robot Control

arXiv:2601.20701v2 Announce Type: replace Abstract: Real-time robot control demands fast action generation. Diffusion and flow matching policies for robot control require multi-step sampling, limiting

On the Off-Policy Teacher in On-Policy Distillation

arXiv:2609.38360v1 Announce Type: cross Abstract: On-policy distillation (OPD) has recently emerged as a promising post-training paradigm in which the student learns from trajectories generated by its

OP-CAD: On-Policy Clean-Audio Distillation for Robust Audio-Visual Reasoning

arXiv:2609.39150v1 Announce Type: new Abstract: Omni-modal large language models deployed in real-world environments encounter external noise that can interfere with their perception and understanding