safety

OpenAI says it 'parted ways' with three researchers for violating its 'handling sensitive' info policies; sources: they shared it with an AI safety organization (Wall Street Journal)

Wall Street Journal: OpenAI says it “parted ways” with three researchers for violating its “handling sensitive” info policies; sources: they shared it with an AI safety organization — OpenAI has parte

safetytechmeme

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

OpenJev-RLCD: A Working RLCD Implementation

DGX agent

arXiv:2609.38850v1 Announce Type: new Abstract: Decision models such as Jev answer questions with probabilities, which are only useful if they are calibrated. Open-source reproductions rely on supervi

safetyarxiv-cs-ai

1 Oct 2026

OPSRD: On-Policy Self-Role Distillation

arXiv:2609.39884v1 Announce Type: cross Abstract: Role prompting elicits specialized behavior from large language models through an expert identity, offering a lightweight way to guide reasoning on de

Optimal Design for Active Preference Learning with Biased LLM Judges

arXiv:2609.38860v1 Announce Type: cross Abstract: Learning from human preferences is central to large language model (LLM) alignment, but human preference annotation is costly. Active preference learn

Optimize, Learn, Refine: Whole-Body Grasping and Pick-and-Throw with a Spiral Soft Robot

arXiv:2609.38202v1 Announce Type: new Abstract: Soft continuum robots can exploit distributed compliance for whole-body manipulation, but synthesizing behavior through changing contacts remains diffic

OPTS-TTPO: Enhancing Finite-Sample Policy-Gradient Learning with Tree Search

arXiv:2609.40035v1 Announce Type: new Abstract: The policy-gradient theorem gives the exact gradient under the current policy, but finite on-policy samples may miss rare high-return trajectories. We s

Outsourcing AI regulation to a voluntary 'pledge' conceived by Mark Zuckerberg and endorsed by tech CEOs over a private White House lunch is…

Outsourcing AI regulation to a voluntary 'pledge' conceived by Mark Zuckerberg and endorsed by tech CEOs over a private White House lunch is a recipe for disaster. 'Self-regulation' by tech giants is

safetygary-marcus--x

Over-Personalization Is a Decision Failure: Generation-Induced Apply Bias in LLMs

arXiv:2609.34284v2 Announce Type: replace Abstract: Personalized LLMs must decide, for each stored preference, whether the current context calls for applying or suppressing it, which we call its appli

PhasePlan: Ordered Future-Phase Planning for Robot Brain Models

arXiv:2609.39970v1 Announce Type: new Abstract: Robot brain models integrate vision, language, and robot state to generate actions for complex manipulation tasks. Most predict fixed-length action chun

PhaseSync-Exo: Human Clock Anchored Reference Adaptation for Dynamic Gait Tracking

arXiv:2609.38824v1 Announce Type: new Abstract: Human-aware exoskeleton walking requires reconstructing gait, tracking diverse motions under dynamic constraints, and preserving human timing. We presen

Policy-Conditioned AI-Use Detection: An Evidentiary Framework for Academic Publishing

arXiv:2609.38427v1 Announce Type: cross Abstract: Major venues now publish detailed rules about how authors, reviewers, and area chairs may use AI, and those rules differ by role, by task, and by what

Policy Iteration Is Not Strongly Polynomial for Deterministic Markov Decision Processes: The Price of Algorithmic Anarchy

arXiv:2609.40147v1 Announce Type: new Abstract: We establish an exponential iteration lower bound in the number of states for Howard's policy iteration on deterministic discounted Markov decision proc

Prediction is Better than Detection: Traffic Congestion Control using Drones

arXiv:2609.39637v1 Announce Type: new Abstract: A central question in deploying teams of mobile robots for persistent monitoring is how task performance scales with fleet size, and whether this scalin

PrefPI: Preference-Guided Steering into Out-of-Distribution Behaviors

arXiv:2609.40165v1 Announce Type: cross Abstract: We present PrefPI (Preference-Guided Policy Iteration), an iterative framework for steering pretrained generative robot policies using only relative p

Project and Mix: Task-Semantic Prototypes for Few-Shot Image Classification

arXiv:2603.24528v2 Announce Type: replace Abstract: Vision-language models like CLIP are trained with the objective of aligning text and image pairs. Beyond text prompts alone, recent works show that

Prototype-guided Bilateral Alignment Multimodal Federated Learning

arXiv:2609.38925v1 Announce Type: new Abstract: Multimodal federated learning (MFL) has emerged as a pivotal paradigm for leveraging distributed data to enhance model performance. However, existing me

Prototype-Rule Neurosymbolic Regularization for Rank-Constrained Tensor Neural Networks under Label Scarcity

arXiv:2609.40131v1 Announce Type: cross Abstract: Rank-constrained tensor neural networks reduce the parameterization of high-order inputs, but they do not explicitly constrain class geometry in the l

Proximal Balancing for Causal Effect Estimation under Unmeasured Confounding

arXiv:2609.40051v1 Announce Type: cross Abstract: Estimating causal effects from observational data is central to science and policy, but the effects are not identified when confounders are unmeasured

Recovering Off-Policy Supervision for Speculative Decoding

arXiv:2609.38795v1 Announce Type: new Abstract: Block drafters for speculative decoding are commonly trained on corpora written by external models, where a single off-policy token invalidates supervis

ReDiF: Resource-Efficient Few-Step Diffusion Distillation via Reinforcement Learning

arXiv:2512.22802v2 Announce Type: replace-cross Abstract: Step distillation accelerates diffusion sampling by training a few-step student to imitate a many-step teacher, but distillation itself remain

Refusals That Bend: Measuring and Predicting Task Malleability in Embodied VLM Planners

arXiv:2609.38971v1 Announce Type: cross Abstract: Embodied vision-language models (VLMs) are increasingly deployed as high-level planners for robots because they generalize across diverse environments

ReGain: Restoring Subject Fidelity in Personalization on Synthetic Images

arXiv:2609.38680v1 Announce Type: cross Abstract: Text-to-image diffusion models are personalized to a subject by DreamBooth fine-tuning on a handful of its images. Increasingly, these images come fro

Reinforcement Learning with Complex (valued) Memories

arXiv:2609.38598v1 Announce Type: new Abstract: Partially observable environments pose a fundamental challenge in deep reinforcement learning, requiring agents to compress temporal information from ob

Report: Anthropic targets pre-Thanksgiving IPO launch, despite warning of AI’s ‘existential risks’

Artificial intelligence giant Anthropic PBC is reportedly forging ahead with its plans to go public and has set a new target date of Nov. 9, according to a report from Bloomberg that cited anonymous s

safetysiliconangle

Reserve-Aware Contrast Certificates for Conservative Bandits with Uncertain Baselines

arXiv:2609.39106v1 Announce Type: new Abstract: Conservative bandits must improve an incumbent policy without exhausting a prescribed performance budget. When the incumbent is uncertain, separately bo

ReTaCo: Residual-Target Control for On-Policy Distillation

arXiv:2609.39275v1 Announce Type: new Abstract: On-policy distillation (OPD) trains a student on its own generated prefixes with token-level teacher feedback, but transmitting or storing the teacher's

Rethinking Legibility in Social Robot Hallway Navigation: Impact of Intent Representation and Human Distraction

arXiv:2609.40158v1 Announce Type: new Abstract: We focus on legible robot motion generation in social navigation settings. Legibility in human-robot interaction (HRI) is often described as the propert

Revisiting On-policy Adversarial Black-Box Distillation: Calibrating Groupwise Reward Geometry for Effective Advantage Construction

arXiv:2609.39757v1 Announce Type: cross Abstract: Black-box distillation is a practical route for transferring capabilities from API-accessible large language models that expose only text outputs into

Revisiting scaling laws for reward optimization

arXiv:2609.38526v1 Announce Type: cross Abstract: Scaling laws for optimization against reward models in AI alignment have pinned down how performance depends on optimization effort---measured by a KL

ReWAM: Reciprocal World Action Models for Interactive Autonomous Driving

arXiv:2609.39245v1 Announce Type: new Abstract: In interactive scenarios, an autonomous driving system is required to generate ego actions under the influence of other agents' behaviors. Existing Worl

Risk-Aware Adaptive Evaluation: Finding High-Impact Failures Under Limited Budgets

arXiv:2609.38914v1 Announce Type: new Abstract: Evaluating interactive agents is expensive. Agent behavior is stochastic, so reliability must be measured over repeated trials, but failures are rare an

RL-Guided PAC-NMPC for Probabilistically-Safe Perception-Based Navigation in Unknown Environments

arXiv:2609.39854v1 Announce Type: new Abstract: In this paper, we present an approach for combining stochastic nonlinear model predictive control (SNMPC) and reinforcement learning (RL) to enable prob

RLTL;DR: Self-Improvement by Internalizing Self-Generated Feedback

The common paradigm of reinforcement learning with verifiable rewards (RLVR) is to let agents make multiple attempts at a task, and optimize towards the successful ones. This becomes problematic in th

safetyapple-ml-research

RoboAssist: Interactive Human-Humanoid Planning for Long-Horizon Surgical Assistance

arXiv:2609.39384v1 Announce Type: new Abstract: Long-horizon surgical assistance requires humanoid robots to coordinate with evolving human activities while maintaining safety across planning and exec

RoboCoach: World Models as Active Coaches for Compositional Robot Skills

arXiv:2609.39685v1 Announce Type: cross Abstract: Long-horizon robot manipulation reuses skills across many task compositions, but improving these compositions with additional end-to-end demonstration

Role-Adaptive Policy Optimization for Offline Reinforcement Learning

arXiv:2609.40149v1 Announce Type: new Abstract: Policy regularization in offline reinforcement learning balances policy improvement against reliance on uncertain value estimates. This balance can diff

Safety Monitors Mostly Catch What the Model Already Refuses

arXiv:2609.05797v3 Announce Type: replace Abstract: Safety monitors are evaluated by recall on harmful prompts, regardless of whether the target model would answer them. Yet a monitor matters most on

Safety of Latent Communication in Multi-Agent Systems

arXiv:2609.39788v1 Announce Type: new Abstract: Latent communication enables multi-agent systems to exchange information directly in internal representation space, reducing the token, computation, and

Scale and Selection: What Makes Automatic Harness Evolution Work for Visual-Interface Robot Agents

arXiv:2609.39304v1 Announce Type: cross Abstract: When an off-the-shelf coding agent is used directly as a robot policy, observing a browser-based 3D interface through screenshots and acting by posing

Semantic Misalignment in Vision-Language Models under Perceptual Degradation

arXiv:2601.08355v3 Announce Type: replace Abstract: Vision-Language Models (VLMs) are increasingly deployed in autonomous driving and embodied AI systems, where reliable perception is critical for saf

Semifactual Credit-Augmented Policy Optimization

arXiv:2609.40360v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has improved the reasoning capabilities of large language models (LLMs), yet their predictions r

Shared Weights, Selected Computations: How Looped Transformers Route What Each Loop Does

arXiv:2609.39892v1 Announce Type: new Abstract: Looped Transformers repeatedly apply the same set of Transformer layers, giving them a recurrent architecture for latent computation. Their strong perfo

ShieldCLIP: Selective Safety Alignment for Harmful Content Mitigation in Multimodal Foundation Models

arXiv:2609.39688v1 Announce Type: cross Abstract: Multimodal encoders such as CLIP underlie many downstream systems, but their web-scale training data embed harmful associations that safety alignment

SimEX: Simulation-Integrated Robotics AutoResearch

arXiv:2609.38982v1 Announce Type: cross Abstract: Coding agents powered by large language models (LLMs) have shown remarkable abilities to autonomously reason about and achieve goals in the digital wo

Smaller Models, Better Rejects: Preference Distillation Scaling

arXiv:2609.38987v1 Announce Type: new Abstract: Preference distillation typically treats a teacher response as preferred and the student's own response as rejected. This assumes that self-generated fa

Social-WM: Safety-Aware Latent World Models for Robot Social Navigation

arXiv:2609.40177v1 Announce Type: new Abstract: Safe social navigation requires a robot to anticipate not only the future consequences of its actions, but also whether a nominal action can actually be

Soft Spatial Reasoning

arXiv:2609.38717v1 Announce Type: cross Abstract: Large Vision-Language Models (LVLMs) commonly perform spatial reasoning through chain-of-thought (CoT), encoding intermediate reasoning as autoregress

Speculative Safety Honeypot: Toward Proactive Defense Against Multi-turn Agent Attacks

arXiv:2609.39549v1 Announce Type: cross Abstract: As Large Language Model (LLM) agents are increasingly deployed in complex environments, multi-turn interaction attacks have become a significant secur