safety

AgentFold: Closed-Loop Agentic Search for Protein Folding Model Design

arXiv:2608.26747v1 Announce Type: new Abstract: Scientific LLM agents have shown promise in literature reasoning, tool use, and experiment planning, but it remains unclear whether they can autonomousl

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

AirLLM: Diffusion Policy-based Adaptive LoRA for Remote Fine-Tuning of LLM over the Air

DGX agent

arXiv:2507.11515v2 Announce Type: replace-cross Abstract: Operating Large Language Models (LLMs) on edge devices is increasingly challenged by limited communication bandwidth and strained computationa

safetyarxiv-cs-ai

28 Aug 2026

Approved Too Late: Verdict Staleness in LLM-Guarded Self-Adaptive Systems

arXiv:2608.26306v1 Announce Type: new Abstract: A large language model (LLM) guardrail for a self-adaptive system (SAS) may issue an approval that is correct at check time but stale by actuation. This

Arrive and Survive: Scaling Safe Goal-Conditioned Policy Learning from One-Bit Failure Signals

arXiv:2608.26571v1 Announce Type: new Abstract: Contrastive reinforcement learning (CRL) scales effectively in goal-conditioned tasks by casting policy learning into a self-supervised contrastive obje

Assessing the Downstream Utility of Evidence-Aware Retrieval in RAG

arXiv:2608.26379v1 Announce Type: cross Abstract: Retrieval evaluation for retrieval-augmented generation (RAG) is increasingly designed around whether retrieved passages contain evidence that can sup

Barrier Function Conformal Safety Clearance Certification with CVaR for Driving Trajectory Selection

arXiv:2608.26533v1 Announce Type: cross Abstract: Autonomous driving motion planners generate and select candidate trajectories while accounting for interactions with surrounding agents. However, thes

Beyond Atomic Layouts: Compositional Design Understanding with Vision-Language Models

arXiv:2608.26716v1 Announce Type: new Abstract: Layout understanding, or the interpretation of element organization, is essential for document analysis, user interface (UI) creation, and graphic desig

Beyond Client Averaging: A Client-Independent Second-Order Stationary-Bias Component in Stochastic SCAFFOLD

arXiv:2608.26765v1 Announce Type: new Abstract: Existing constant-step analysis of stochastic Scaf{} identifies a leading O(gamma/N) stationary mean bias and shows that higher-order bias can persist a

Beyond the Proving Ground: Independent Public-Road Testing of Assisted Lane Change Systems using LiDAR

arXiv:2608.26669v1 Announce Type: cross Abstract: Testing of commercial Advanced Driver Assistance Systems is essential to ensure safety and compliance during type approval and in service operation. H

Boosting LLM Exploration via Weak-Model Guidance in RLVR

arXiv:2608.27420v1 Announce Type: new Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) significantly improves LLM reasoning but often causes a drop in policy entropy, leading to narrowe

CARE: Causally-Aligned Reasoning Exploration for Medical Large Language Models

arXiv:2608.26147v1 Announce Type: new Abstract: Large Language Models (LLMs) have shown strong potential for medical reasoning, yet the scarcity and cost of expert-annotated data constrain their progr

Case2Flow: Bridging Patient Cases and Guideline Flowcharts through Multimodal Retrieval

arXiv:2608.26414v1 Announce Type: new Abstract: Medical guidelines encode rich, evidence-based decision logic, yet the specific decision artifact a clinician needs is hard to locate within a guideline

Categorizer Automata for Discounted-Sum Payoffs

arXiv:2608.26763v1 Announce Type: new Abstract: Categorizing continuous data into discrete bins is a fundamental operation in artificial intelligence. We introduce the categorizer automaton, a determi

CLIPPER: Replayable Shortlisted Optimization for Repeated Spatial Coverage Planning

arXiv:2608.26819v1 Announce Type: new Abstract: Operational requirements developed with the City of Braunschweig frame municipal micromobility planning under geofenced exclusions, mandatory retained s

Communication styles and reader preferences of LLM- and human-authored COVID-19 information explanations: a case study

arXiv:2505.08143v2 Announce Type: replace-cross Abstract: With the wide adoption of large language models (LLMs) in information assistance, it is essential to examine their alignment with human commun

Counterfactual Bias Testing for Application Tracking System

arXiv:2608.26899v1 Announce Type: new Abstract: Automated candidate-job matching systems are increasingly classified as high-risk AI under emerging regulation, yet auditing them for demographic bias i

Cross-Lingual Alignment Without Joint Training: Do Monolingual Language Models Converge on Universal Representations?

arXiv:2608.27115v1 Announce Type: new Abstract: Cross-lingual alignment in multilingual language models is typically attributed to joint training: shared parameters, mixed-language batches, or explici

Cross-Platform Generalisation Failure in Mental Health Natural Language Processing: A Five-Axis Fairness Audit of Transformer Models on Social Media

arXiv:2608.26138v1 Announce Type: new Abstract: We introduce the Cross-Platform Fairness Evaluation (CPFE) framework -- a five-axis audit protocol covering discriminative performance, calibration, sta

Decoupling Planning and Control for Instructable Agents

arXiv:2608.26788v1 Announce Type: new Abstract: Recent work shows that pre-trained, instruction-tuned vision-language models (VLMs) perform well at mapping from instructions and observations to high-l

Difference-in-Differences on a Censored Rating Scale Can Manufacture an Effect: Evidence from a Pre-Registered LLM-Judge Audit

arXiv:2608.27309v1 Announce Type: cross Abstract: Audits of LLM judges certify a bias by contrasting matched conditions, and the strongest designs difference twice: a within-item contrast between two

Disentangling Optimization Scale from Preference Scale in DPO

arXiv:2608.27032v1 Announce Type: new Abstract: Direct Preference Optimization (DPO) is a widely used objective for aligning language models from preference data, with the coefficient eta commonly int

DPA-I2P: Depth-Guided Projective Alignment for Image-to-Point-Cloud Registration in Autonomous Driving

arXiv:2608.26589v1 Announce Type: new Abstract: Image-to-Point Cloud Registration aims to estimate the camera pose of a given image within a 3D scene point cloud, which is a fundamental task in autono

Egosurg: Arbitrary view synthesis for egocentric replay of operating room workflows from ambient cameras

arXiv:2510.04802v2 Announce Type: replace-cross Abstract: Observing surgical practice has historically relied on fixed vantage points or recollections, leaving the egocentric perspectives that shape c

Emotional Preferences as Goal-Priority Regulation

arXiv:2608.27072v1 Announce Type: cross Abstract: A core question in decision-making for agents is whether the relative priorities of competing lower-level objectives can be determined by emotional pr

Evaluating AI Generated Summaries for Cancer Patients

arXiv:2608.26154v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly being integrated into digital health platforms to generate summaries of complex medical data. Although t

Evaluator-Dependent Patient-Adaptive ECG Lead-Channel Allocation

arXiv:2608.26827v1 Announce Type: new Abstract: Patient-conditioned acquisition policies for ECG lead-channel selection can outperform population-wide fixed protocols by tailoring the channel budget t

Five Primitives for Governing Autonomous AI Agents at Runtime

arXiv:2608.26696v1 Announce Type: new Abstract: Enterprise deployments of autonomous AI agents inherit a control model built for human users and long-lived services, and the fit fails in three specifi

FLARE: A Failure-Aware Framework for Autonomous Correction and Recovery in Visual-Language Robotic Manipulation

arXiv:2608.26645v1 Announce Type: new Abstract: Vision-Language-Action Models~(VLAs) have demonstrated significant promise in generalizing to complex, long-horizon robotic manipulation tasks. However,

Gary and I co-wrote this piece about the security standards we should expect from the AI labs. It’s our attempt to bring some clarity to the…

Gary and I co-wrote this piece about the security standards we should expect from the AI labs. It’s our attempt to bring some clarity to the OpenAI Hugging Face incident. Now live (and free) at Marcus

safetygary-marcus--x

GeoMAD: Geometry-Aware Multi-View Anomaly Detection via Deformable Fusion and Distributional Alignment

arXiv:2608.26724v1 Announce Type: new Abstract: Multi-view anomaly detection (MvAD) detects defects by exploiting complementary observations from multiple camera viewpoints. The central challenge is t

GRAFT: Grounded and Efficient Online Reinforcement Adaptation for Fine-Grained Robot Manipulation

arXiv:2608.27079v1 Announce Type: new Abstract: Pretrained vision-language-action (VLA) policies provide strong priors for robot manipulation, yet adapting them online to fine-grained biomedical tasks

GRAS: Guided Reduced-Variance Proposals and Adaptive Selection for Training-Free Reward Alignment in Discrete Diffusion

arXiv:2608.26585v1 Announce Type: new Abstract: Discrete diffusion models have become a strong, widely adopted class of generators for sequence data, and steering them toward a downstream reward at in

Gromov-Monge Flow Matching for Equivariant Graph Generation

arXiv:2608.26961v1 Announce Type: new Abstract: Graphs are invariant under node permutations, motivating the use of permutation-equivariant architectures in generative models. In flow matching, howeve

Incremental Recommendation via Causal Models

arXiv:2608.26804v1 Announce Type: cross Abstract: Recommendation impressions are a finite resource, hence delivering a recommendation to a user who would discover the content organically yields no inc

Instruction Quality Matters: Refining Instructions for Effective Preference Learning

arXiv:2608.26779v1 Announce Type: new Abstract: Preference learning optimizes models using response pairs, yet the informativeness of these pairs is fundamentally shaped by the instructions from which

INTENT-AS-A-TOOL Makes it Easy to Track Agentic Misalignment

arXiv:2608.27348v1 Announce Type: new Abstract: As large language models (LLMs) are deployed as autonomous agents, safety failures increasingly involve consequential actions. We study agentic misalign

Interpretable, Fairly Evaluated Automated L2 Speaking Assessment that Beats the Single-Human Ceiling and Why Pause Encoding Does Not Change LLM Fluency Scores

arXiv:2608.26137v1 Announce Type: new Abstract: Second-language (L2) English learners can rarely rehearse speaking with a partner. Speaking is also the most anxiety-laden skill. These gaps drive a fas

Is Your Neighborhood Safe? Place-based Stigma in Large Language Models' Urban Safety Judgments

arXiv:2608.26188v1 Announce Type: new Abstract: Large language models are increasingly used to inform safety decisions in cities, such as where it is safe to walk, rent, or travel. We ask whether such

ITL: Interpretable Document Alignment with Structured Reference Frameworks

arXiv:2608.27031v1 Announce Type: new Abstract: Measuring alignment between documents and structured reference frameworks requires identifying conceptual evidence distributed throughout the text and r

just had a quick look at microduck_rl the codebase is very elegant, i’d recommend reading the agent.md since it contains quite a few fun qui…

just had a quick look at microduck_rl the codebase is very elegant, i’d recommend reading the agent.md since it contains quite a few fun quirks for reward modeling like head tracking too tight impairs

safetyclem-delangue--x

LAAF: A Layered Accountability Architecture Framework for LLM Applications

arXiv:2608.27102v1 Announce Type: new Abstract: Large Language Models (LLMs) operate in hospitals, courtrooms, banks, and public service desks, where fluent, confident outputs are treated as authorita

Learning-Augmented Online Allocation under Unreliable Advice: Robustness, Exposure Fairness, and Distribution Shift

arXiv:2608.26889v1 Announce Type: new Abstract: Learning-augmented algorithms improve online decisions using predictions, but unreliable advice may harm efficiency and fairness. We study an online all

Learning to Predict, Discover, and Reason in High-Dimensional Event Sequences

arXiv:2603.16313v3 Announce Type: replace Abstract: Electronic control units (ECUs) embedded within modern vehicles generate a large number of asynchronous events known as diagnostic trouble codes (DT

Learning Woody Clearing With Loss Alignment for Zero-Shot Regrowth and Woody Segmentation

arXiv:2608.26489v1 Announce Type: new Abstract: Detecting woody clearing is vital for managing biodiversity. Deep learning models can detect change in woody vegetation from bitemporal remote sensing i

Making Latent Evolution Explicit: Operator-Structured Transitions for World Action Models

arXiv:2608.27259v1 Announce Type: new Abstract: World Action Models (WAMs) augment robot policies by predicting how task-relevant scene states may evolve under interaction. Recent WAMs increasingly pe

MedFG-VQA: Low-Frequency Memory and Graph Attention for Lightweight Medical VQA

arXiv:2608.26848v1 Announce Type: cross Abstract: Medical Visual Question Answering (Med-VQA) holds significant promise for clinical decision support, yet faces challenges due to limited annotated dat

MENTOR: Reinforcement Learning via Flexible Teacher-Optimized Rewards for Tool-Use Distillation

arXiv:2510.18383v4 Announce Type: replace-cross Abstract: Distilling the tool-use capabilities of large language models (LLMs) into small language models (SLMs) is essential for their practical applic

Mitigating Fabrication in Multi-Stage LLM Pipelines for Hiring: An Empirical Evaluation of Prompt Guardrails and Human-in-the-Loop Checkpoints

arXiv:2608.26171v1 Announce Type: cross Abstract: Multi-stage LLM hiring pipelines (resume improvement, interview question generation, answer feedback) can fabricate credentials, inflate qualifiers, a