Search: “safety”

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

Reserve-Aware Contrast Certificates for Conservative Bandits with Uncertain Baselines

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

DGX agent

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

safetyarxiv-cs-ai

1 Oct 2026

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

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

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

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

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

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

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

Staying on Task: Testing the Foundations of Long-Horizon Agent Reliability

arXiv:2609.38712v1 Announce Type: new Abstract: Long-horizon agentic workflows require models to sustain repeated state-dependent actions all while the context grows, sub-task complexity changes, and

Steepest Guidance: A Practical and Principled Approach to Inference-Time Alignment of Flow and Diffusion-based Models

arXiv:2609.39091v1 Announce Type: cross Abstract: Inference-time alignment of flow and diffusion-based models is critical for achieving flexible generative modeling. Theoretically, Doob's h-transform

Stress-Testing LLM Lie Detectors: Role-Play Failures and Spurious Correlations

arXiv:2609.39807v1 Announce Type: cross Abstract: Lie detection probes aim to predict from a language model's internal states whether its output is truthful or dishonest. However, role-play complicate

Strike a Chord! Modal Kinetic Typography

arXiv:2609.38325v1 Announce Type: new Abstract: We introduce modal kinetic typography, which animates a vector glyph to express a semantic concept while keeping it legible. Our key idea is to build mo

TERRA: Terrain-Aware Reconstruction, Retargeting and Control for Musculoskeletal Locomotion

arXiv:2609.38653v1 Announce Type: cross Abstract: Recent advances in musculoskeletal modeling and reinforcement learning have enabled muscle-actuated agents to reproduce increasingly complex human mot

Text-to-3D Policy: Fine-Grained Language-Behavior Alignment for Unseen Specification Generalization

arXiv:2609.39599v1 Announce Type: cross Abstract: 3D visuomotor policies provide a strong foundation for spatially precise manipulation, yet current text-to-3D policies struggle to follow unseen fine-

The Missing Coefficients: Bayesian Pairwise Merging for Model Personalization

arXiv:2609.39055v1 Announce Type: new Abstract: How can we personalize a shared expert library from a user's pairwise choices? Prior work can realize different reward trade-offs by merging reward-spec

The Planning Limits of Latent World Models

arXiv:2609.39235v1 Announce Type: cross Abstract: World models offer a promising way to help robots understand how the physical world evolves and plan complex behaviours through imagination. Yet exist

Tool-Policy Co-Design for Powder Weighing in Laboratory Automation

arXiv:2609.39797v1 Announce Type: new Abstract: Autonomous powder weighing is one of many bottlenecks in laboratory automation due to the complex, non-linear dynamics of heterogeneous materials. Robot

Towards Optimal Inventory Control under Censored Demand: A Biased Sample-Average Approximation Approach

arXiv:2609.39397v1 Announce Type: cross Abstract: We study data-driven multi-period lost-sales inventory control under censored demand, where a stockout reveals only that demand exceeded the stocking

Towards Universal Wasserstein Barycenters through Flow Matching

arXiv:2609.38547v1 Announce Type: cross Abstract: Defining a weighted mean over probability measures under probability metrics is a central tool in probabilistic machine learning. Under the Wasserstei

Training-Free Affinity Fusion of Neural and Embedding-Based Speaker Diarization

arXiv:2609.39162v1 Announce Type: cross Abstract: Speaker diarization systems based on speaker embeddings and neural diarization exploit complementary forms of speaker information, but their intermedi

Understanding Off- vs On-Policy Distillation: A Tale of Distinct Training Objectives

arXiv:2609.38666v1 Announce Type: cross Abstract: On-policy distillation (OPD) learns from teacher feedback on student-generated responses and has shown promise in reducing forgetting relative to supe

Unveiling the Value of Motion for Cinematic Camera Trajectories

arXiv:2609.38683v1 Announce Type: cross Abstract: Cinematic camera motion is a fundamental storytelling tool, defined not only by where the camera is positioned in the scene, but also by how it moves

VERA: Verifiable Feasibility Representations with Counterfactual Credit for Constrained Multi-Agent Control

arXiv:2609.38889v1 Announce Type: new Abstract: Constrained multi-agent control requires more than predicting rewarding actions: an action can cease to be executable as contact windows, shared capacit

WEIRDO: WEak resIdual Regularized DOob's h-transform diffusion alignment

arXiv:2609.39531v1 Announce Type: cross Abstract: We study the problem of estimating the guidance that steers the distribution learned by a diffusion generative model toward a tilted target q_0 propto

What to Attend, What to Keep: Skill-Conditioned Visuotactile Representation with Progress-Guided Event Memory

arXiv:2609.38494v1 Announce Type: cross Abstract: Robotic manipulation integrates vision, touch, and language, whose importance shifts across stages: vision guides reaching, while touch, through its e

When Clipping Reverses Correction: Failure Dynamics of Pointwise Forward-KL On-Policy Self-Distillation

arXiv:2609.38995v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) trains a student on its own generated responses using feedback from the same model conditioned on privileged informat

safetyarxiv-cs-cl

When Instructions Retrieve Trajectories: Diagnosing and Mitigating Generalization Failures in VLA Models

arXiv:2609.39971v1 Announce Type: cross Abstract: Vision-language-action (VLA) models can exceed 90% success on in-distribution tasks and withstand nuisance changes that preserve the required action,

When Order Matters: First-Speaker Bias and Mitigation through Personality in Sequential Multi-Agent Debate

arXiv:2609.38964v1 Announce Type: new Abstract: Multi-agent debate (MAD) is often used to improve large language model (LLM) reasoning, but sequential debate is rarely a neutral aggregator of agents'

Whose Voice Survives the Summary? A Voice-Retention Audit of LLM Employee Listening

arXiv:2609.38818v1 Announce Type: new Abstract: Organizations increasingly route employee feedback to leaders through large language model (LLM) summaries, an unaudited layer that silences already-spo

World-as-Graph: Relational World Modeling Through Latent Space Graphs

arXiv:2609.38927v1 Announce Type: cross Abstract: World models aim to learn representations of real-world environments and predict their future evolution. Recent object-centric world models have made

Zero-Compute Cross-Lingual Transferability Estimation Using Typological Feature Proxies

arXiv:2609.39640v1 Announce Type: cross Abstract: Cross-lingual transfer describes how knowledge in a source language benefits a target language. Measuring it quantitatively requires broad multilingua

A Comprehensive View of Fairness through Distributional Stability

arXiv:2609.37061v1 Announce Type: cross Abstract: We view fairness as a property of distributional stability. Rather than assessing a predictor under a fixed data distribution, we study how its predic

A Spectral Theory of Distortion in LLM Graph Reconstruction: Sharp Bounds and Empirical Characterization

arXiv:2609.38161v1 Announce Type: new Abstract: Evaluations of graph reconstruction by language models typically report a single aggregate distance between the original and the reconstructed graph. We

ABC: Advantage-Based Control Variates for Reinforcement Learning with Verifiable Rewards

arXiv:2609.36058v1 Announce Type: new Abstract: Recent progress in reinforcement learning with verifiable rewards (RLVR) has highlighted the effectiveness of simple critic-free policy-gradient methods

Achieving an O(1/N) Optimality Gap in Average-Reward Weakly-Coupled MDPs

arXiv:2609.38132v1 Announce Type: new Abstract: We study average-reward weakly-coupled Markov decision processes (WCMDPs), where a WCMDP consists of N smaller MDPs, called arms, that share multiple pe

Action Chunking Proximal Policy Optimization with Feedback Correction

arXiv:2609.36250v1 Announce Type: cross Abstract: Action chunking provides temporal abstraction in reinforcement learning by selecting short action sequences instead of individual actions, but many ex

AdaTutoRank: Learning to Rerank Document Sets via Adaptive Tutoring Optimization for RAG and Deep Research

arXiv:2609.32472v2 Announce Type: replace Abstract: Document rerankers determine what evidence reaches the downstream model in RAG and deep research, yet mainstream rerankers select by relevance match

safetyarxiv-cs-cl

Adversarial Training for Pixel Diffusion

arXiv:2609.38170v1 Announce Type: new Abstract: Pixel diffusion models generate RGB images directly, avoiding the bottleneck of an autoencoder, yet their outputs still systematically underrepresent fi

All You Need Is Low Fidelity: Zero-Shot Sim-to-Real of Learned Robotic Fish Control

arXiv:2609.36993v1 Announce Type: new Abstract: Complex tasks for underwater robots remain limited by the capabilities of their controllers. Learning a better one for a soft, underactuated robotic fis

Altered Thoughts, Altered Actions: Reasoning Chain as Control Surface for a Vision-Language-Action Policy

arXiv:2603.12717v2 Announce Type: replace-cross Abstract: Vision-language-action policies map camera images and natural-language instructions to a robot's motor actions. Some of these policies are des

An Empirical Study and Assessment of EU AI Act Compliance Checkers

arXiv:2609.36228v1 Announce Type: new Abstract: The EU AI Act introduces extensive compliance requirements for organizations that develop, deploy, or integrate AI systems. Many of these requirements a

Announcing our first open-weights world-action model, Praxis-1. It turns out that, unsurprisingly, you can get predictably better robotics p…

safetycristobal-valenzuela--x

Announcing our first open-weights world-action model, Praxis-1. It turns out that, unsurprisingly, you can get predictably better robotics policy performance by scaling third-person video. It’s been f