AI safety and alignment

Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.

18,070 stories · RSS feed

arXiv Machine Learning

Jul 21

The Concept of Representation in ML: Beyond Plato and Aristotle

arXiv:2607. 17800v1 Announce Type: new Abstract: Representation is a central concept in modern machine learning, where it usually refers to internal encodings that support learning and generalization.

By Gilad Landau, Aviv Keren

safety

arXiv Machine Learning

Jul 21

fSRD: Fuzzy Spectral Region Decomposition -- Automated Multi Operator Koopman Representations via an Adaptive Spectral Learning Architecture

arXiv:2607. 17990v1 Announce Type: new Abstract: Highly nonlinear chaotic dynamical systems remain difficult to model due to fundamental trade-offs between complexity, expressivity, and data efficiency.

By Charles Bokor, Mark Cary, Denise Morrey, Fabrizio Bonatesta

rag safety

arXiv Machine Learning

Jul 21

What does a Bayes-filtered transformer believe? A predictive Monte Carlo approach

arXiv:2607. 17060v1 Announce Type: new Abstract: A Bayes-filtered transformer (BFT) is a transformer trained on sequences that are generated in two steps: first a latent task is drawn from a prior, then observations are drawn conditional on that task.

By Afiq Abdillah Effiezal Aswadi, Haotong Ma, Susan Wei

llms diffusion safety

arXiv Machine Learning

Jul 21

When Can Safe Controllers Adapt? Information before Commitment

arXiv:2607. 16895v1 Announce Type: new Abstract: Safe adaptive control is online adaptation under a safety guarantee on the learning trajectory itself.

By Venkatesh Saligrama

safety

arXiv Machine Learning

Jul 21

Preference-based Antibody Expression Ranking: Scaling with Large-scale Weak Supervision

arXiv:2607. 16263v1 Announce Type: new Abstract: Antibody expression ranking is a critical task in antibody design, yet its modelling is severely hindered by the scarcity of labeled experimental data.

By Josh Qixuan Sun, Morteza Babaie, Wenyang Hou, Mark Crowley, David Young

llms reinforcement-learning safety

arXiv Machine Learning

Jul 21

Self-Evolving Just-In-Time Memory for Proactive Embodied Safety

arXiv:2607. 16247v1 Announce Type: new Abstract: While Vision-Language Models (VLMs) have empowered embodied agents to execute complex household tasks, they struggle to proactively handle dynamically emerging hazards during closed-loop interactions.

By Bingrui Sima, Lizhong Wang, Xiaoya Lu, Kun He, Xiao Yang

llms agents robotics multimodal safety

arXiv AI

Jul 21

HiCI: Hierarchical Construction-Integration for Long-Context Attention

arXiv:2603. 20843v3 Announce Type: replace-cross Abstract: Long-context language modeling is commonly framed as a scalability challenge of token-level attention, yet local-to-global information structuring remains largely implicit in existing approaches.

By Xiangyu Zeng, Qi Xu, Yunke Wang, Chang Xu

llms benchmarks safety

arXiv AI

Jul 21

MMGraphRAG: Bridging Vision and Language with Interpretable Multimodal Knowledge Graphs

arXiv:2507. 20804v3 Announce Type: replace Abstract: Large Language Models (LLMs) suffer from hallucinations due to their static parametric knowledge.

By Xueyao Wan, Hang Yu

llms rag multimodal benchmarks safety

arXiv AI

Jul 21

RobustVLA: On Robustness of Vision-Language-Action Model against Multi-Modal Perturbations

arXiv:2510. 00037v5 Announce Type: replace-cross Abstract: In Vision-Language-Actionf(VLA) models, robustness to real-world perturbations is critical for deployment.

By Jianing Guo, Zhenhong Wu, Chang Tu, Yiyao Ma, Xiangqi Kong, Zhiqian Liu, Jiaming Ji, Shuning Zhang, Yuanpei Chen, Kai Chen, Qi Dou, Yaodong Yang, Xianglong Liu, Huijie Zhao, Weifeng Lv, Simin Li

llms diffusion reinforcement-learning robotics multimodal safety

arXiv AI

Jul 21

Li-ViP3D++: Query-Gated Deformable Camera-LiDAR Fusion for End-to-End Perception and Trajectory Prediction

arXiv:2601. 20720v2 Announce Type: replace-cross Abstract: End-to-end perception and trajectory prediction from raw sensor data is one of the key capabilities for autonomous driving.

By Matej Halinkovic, Nina Masarykova, Alexey Vinel, Marek Galinski

agents robotics multimodal safety

arXiv AI

Jul 21

SONAR: Spectral-Contrastive Audio Residuals for Generalizable Deepfake Detection

arXiv:2511. 21325v2 Announce Type: replace-cross Abstract: Deepfake (DF) audio detectors still struggle to generalize to out of distribution inputs.

By Ido Nitzan Hidekel, Gal lifshitz, Khen Cohen, Dan Raviv

rag diffusion benchmarks safety

arXiv AI

Jul 21

Comprehensive Evaluation of Machine Learning for Type 2 Diabetes Risk Prediction: Large-Scale External Validation and Fairness Analysis

arXiv:2607. 16253v1 Announce Type: cross Abstract: Machine learning-based Type 2 diabetes risk prediction models obtain good internal validation results but lose effectiveness in real-world applications due to deficient external testing and fairness assessment.

By Rajveer Singh Pall, Sameer Yadav, Siddharth Bhalerao, Sourabh Sahu, Ritu Ahluwalia, Bhaskar Awadhiya

safety

arXiv AI

Jul 21

Alignment of a Total Automation Economy

arXiv:2607. 17015v1 Announce Type: cross Abstract: We consider economic theory from the perspective of a total automation economy, one with no human involvement in production either in manufacturing or in management.

By David McAllester

agents safety

arXiv AI

Jul 21

DocOCR-Eval: A Correction-Based Framework for OCR Tool Selection Without Ground Truth

arXiv:2607. 16203v1 Announce Type: cross Abstract: Document parsing is a foundational step for document understanding tasks such as visual question answering and key information extraction, as it transforms unstructured scanned images into structured representations by extracting textual, visual, and layout information.

By Zihan Xu, Puzhen Wu, Lawrence Chun Man Lau, Wei Liu, Sirui Li, Yifan Peng, Yihao Ding

llms computer-vision nlp multimodal benchmarks safety

arXiv AI

Jul 21

Interactive Task Alignment as a POMDP

arXiv:2607. 16412v1 Announce Type: new Abstract: Current benchmarks for language models primarily evaluate execution on fully specified tasks.

By Andy Dai, Zexue He, Zhenyu Zhang, Alex Pentland, Jiaxin Pei

llms reinforcement-learning fine-tuning benchmarks safety

arXiv AI

Jul 21

Explainable Lightweight Compact Deep Models for Speech Emotion Recognition

arXiv:2607. 16803v1 Announce Type: cross Abstract: Speech Emotion Recognition (SER) is an important component in a wide range of human-centered applications, including healthcare, customer service, and human-omputer interaction.

By Nelly Elsayed

safety

arXiv AI

Jul 21

DepthART: Scaling Foundation Monocular Depth to Tiny Models

arXiv:2607. 17099v1 Announce Type: cross Abstract: Recent geometric foundation models (e.

By Feng Xue, Wu Chen, Mingshuai Zhao, Guofeng Zhong, Anlong Ming, Haozhe Wang, Dianqiao Lei, Zhaowen Lin, Haiyang Zhang, Nicu Sebe

computer-vision fine-tuning safety

arXiv AI

Jul 21

Specifying the Delegated-Autonomy Boundary: Requirements Engineering for Agentic AI

arXiv:2607. 17225v1 Announce Type: cross Abstract: Agentic AI systems do not just predict or recommend; they plan, maintain state, and act in external environments with varying degrees of autonomy.

By Chetan Arora, Andreas Vogelsang, Abbi Sharma

agents safety

arXiv AI

Jul 21

DecoyFace: Beyond Obfuscation via Controllable and Imperceptible Identity Misdirection for Privacy-Preserving Face Recognition

arXiv:2607. 17504v1 Announce Type: cross Abstract: Split face recognition reduces client-side computation but exposes intermediate features to feature inversion attacks and unauthorized analysis by honest-but-curious (HBC) servers.

By Zhihan Ren, Lijun He, Xinyao Wang, Xinzhu Fu, Fan Li

safety

arXiv Machine Learning

Jul 21

Certified Training for Convolutional Perturbations

arXiv:2607. 18195v1 Announce Type: cross Abstract: Vision models have been found to be susceptible to perturbations such as motion blur induced at runtime by a shaking camera.

By Benedikt Br\"uckner, Alessio Lomuscio

safety

Get the daily briefing