AI safety and alignment

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

11,365 stories · RSS feed

arXiv AI

Jun 6

CangLing-KnowFlow: A Unified Knowledge-and-Flow-fused Agent for Comprehensive Remote Sensing Applications

arXiv:2512. 15231v3 Announce Type: replace Abstract: The automated and intelligent processing of massive remote sensing (RS) datasets is critical in Earth observation (EO).

By Zhengchao Chen, Haoran Wang, Jing Yao, Jianshe Zhang, Pedram Ghamisi, Jun Zhou, Peter M. Atkinson, Bing Zhang

llms rag agents benchmarks safety

arXiv AI

Jun 6

Causal Scaffolding for Physical Reasoning: A Benchmark for Causally-Informed Physical World Understanding in VLMs

arXiv:2606. 05966v1 Announce Type: cross Abstract: Understanding and reasoning about the physical world is the foundation of intelligent behavior, yet state-of-the-art vision-language models (VLMs) still fail at causal physical reasoning, often producing plausible but incorrect answers.

By Tianyi Tang, Zhuoyi Lin, Zeyu Feng, Tianyi Ma, Yew-Soon Ong, Ivor Tsang, Haiyan Yin

llms fine-tuning multimodal benchmarks safety

arXiv AI

Jun 6

Willing but Unable: Separating Refusal from Capability in Code LLMs via Abliteration

arXiv:2606. 05396v1 Announce Type: cross Abstract: Producing a labeled vulnerable code at scale is a recurring obstacle for learning-based vulnerability detection: mined corpora carry substantial label noise, and existing LLM-based augmentation propagates these inaccuracies because it transforms vulnerable seeds rather than synthesising vulnerabilities from a specification.

By Cristina Carleo, Pietro Liguori, Naghmeh Ivaki, Domenico Cotroneo

llms reinforcement-learning safety data-engineering

arXiv AI

Jun 6

Reducing Hallucinations in Complex Question Answering using Simple Graph-based Retrieval-Augmented Generation (long version)

arXiv:2606. 05901v1 Announce Type: cross Abstract: Large language models (LLMs) have fundamentally transformed the landscape of Natural Language Processing.

By Christopher J. Wedge, Joshua Stutter, Danny Dixon, Jacek Ca{\l}a

llms rag agents nlp fine-tuning benchmarks safety

arXiv AI

Jun 6

SoCRATES: Towards Reliable Automated Evaluation of Proactive LLM Mediation across Domains and Socio-cognitive Variations

arXiv:2606. 05563v1 Announce Type: new Abstract: Evaluating LLM mediators remains challenging, as mediation unfolds as a real-time trajectory shaped by disputants' shifting emotions, intentions, and context.

By Taewon Yun, Hyeonseong Park, Jeonghwan Choi, Hayoon Park, Yeeun Choi, Hwanjun Song

llms agents benchmarks safety

arXiv AI

Jun 6

The Mirage of Performance Gains: Why Contrastive Decoding Fails to Mitigate Object Hallucinations in MLLMs?

arXiv:2504. 10020v4 Announce Type: replace-cross Abstract: Contrastive decoding strategies are widely used to reduce object hallucinations in multimodal large language models (MLLMs).

By Hao Yin, Guangzong Si, Zilei Wang

llms multimodal benchmarks safety

arXiv AI

Jun 6

Where Should Knowledge Enter? A Layered Framework for Knowledge Infusion in Multimodal Iterative Generative Mo

arXiv:2606. 06356v1 Announce Type: new Abstract: Multimodal generative models produce fluent outputs but remain unreliable when generation must respect structured, domain-specific, or safety-critical knowledge.

By Renjith Prasad, Chathurangi Shyalika, Anushka Pawar, Amit Sheth

diffusion fine-tuning multimodal safety

arXiv AI

Jun 6

LLM Self-Recognition: Steering and Retrieving Activation Signatures

arXiv:2606. 06315v1 Announce Type: new Abstract: Recent advances in interpretability suggest that large language models (LLMs) implicitly encode signals in their generated text that enable self-recognition of their outputs.

By Thibaud Ardoin, Jonas Sch\"afer, Gerhard Wunder

llms rag safety

arXiv AI

Jun 6

When AI Says It Feels

arXiv:2606. 05734v1 Announce Type: new Abstract: Large language models (LLMs) are generally constrained from expressing feelings through human-preference alignment in post-training processes.

By Shin-nosuke Ishikawa, Seiya Ikeda, Hirotsugu Ohba

llms reinforcement-learning safety

arXiv AI

Jun 6

TinyML-Driven Cybersecurity for Autonomous Spacecraft: Latency-Accuracy Analysis for SPARTA RF and Cyber Threat Detection

arXiv:2606. 05779v1 Announce Type: cross Abstract: Autonomous spacecraft require rapid, lightweight, and reliable onboard detection of cyber-RF threats.

By Van Le, Trevor Tran, Tan Le

agents robotics safety

arXiv AI

Jun 6

Fault tolerance estimation in digital circuits with visualised generative networks

arXiv:2605. 15212v2 Announce Type: replace-cross Abstract: We propose a new numerical method to estimate the fault tolerance of failure modes in digital circuit structures with a generative network sampling technique.

By Sascha Biel, Carl Alexander Gaede, Amiel Glaser, Jan Wolter, Alexej Schelle

diffusion safety

arXiv AI

Jun 6

Insurance of Agentic AI

arXiv:2606. 05449v1 Announce Type: new Abstract: Agentic artificial intelligence (AI) systems are transforming the risk landscape by extending beyond information generation to autonomous planning, tool invocation, decision execution, and persistent modification of digital and physical environments.

By Quanyan Zhu

agents safety

arXiv AI

Jun 6

Learning of Robot Safety Policies via Adversarial Synthetic Scenarios

arXiv:2606. 05952v1 Announce Type: cross Abstract: In this work, we propose an agentic gamification framework for hazard-informed learning of robot safety policies through synthetic scenarios.

By Nikolai Dorofeev, Alexey Odinokov, Rostislav Yavorskiy

rag agents robotics safety

arXiv AI

Jun 6

A Systematic Analysis of Biases in Large Language Models

arXiv:2512. 15792v3 Announce Type: replace-cross Abstract: Large language models (LLMs) have rapidly become indispensable tools for acquiring information and supporting human decision-making.

By Xulang Zhang, Rui Mao, Erik Cambria

llms nlp safety

arXiv AI

Jun 6

Data Flow Control: Data Safety Policies for AI Agents

arXiv:2606. 05679v1 Announce Type: cross Abstract: Agents increasingly generate SQL, orchestrate pipelines, and automate data analysis on behalf of users.

By Charlie Summers, Eugene Wu

agents safety data-engineering

arXiv AI

Jun 6

LatentWave: JEPA Pretraining for Wireless Foundation Models

arXiv:2606. 06373v1 Announce Type: cross Abstract: Wireless foundation models have emerged as a promising alternative to building separate models for each wireless task.

By Ahmed Mohamed, Ahmed Aboulfotouh, Hatem Abou-Zeid

rag diffusion safety

arXiv AI

Jun 6

Synapse: Federated Tool Routing via Typed Compendium Artifacts

arXiv:2602. 00911v2 Announce Type: replace Abstract: The unit of collaboration in federated learning determines what guarantees are even expressible.

By Abhijit Chakraborty, Yash Shah, Vivek Gupta

llms safety

arXiv AI

Jun 6

Beyond Waveform Robustness: Robust Feature-Vocoder Adversarial Attacks on Automatic Speech Recognition

arXiv:2606. 05678v1 Announce Type: cross Abstract: Automatic speech recognition (ASR) systems have become widely used for multilingual speech-to-text transcription.

By Yifan Liao, Zongmin Zhang, Zhen Sun, Yuhui Sun, Xinhu Zheng, Xinlei He

multimodal benchmarks safety

arXiv AI

Jun 6

HypRAG: Hyperbolic Dense Retrieval for Retrieval Augmented Generation

arXiv:2602. 07739v2 Announce Type: replace-cross Abstract: Embedding geometry plays a fundamental role in retrieval quality, yet dense retrievers for retrieval-augmented generation (RAG) remain largely confined to Euclidean space.

By Hiren Madhu, Ngoc Bui, Ali Maatouk, Leandros Tassiulas, Smita Krishnaswamy, Menglin Yang, Sukanta Ganguly, Kiran Srinivasan, Rex Ying

llms rag benchmarks safety

arXiv AI

Jun 6

Whose Alignment? Comparing LLM Process Alignment Across Diverse Organizational Decision Contexts

arXiv:2605. 25256v2 Announce Type: replace Abstract: Steerable pluralism requires a model to faithfully represent one specified perspective.

By Niklas Weller, Emilio Barkett

llms benchmarks safety

Get the daily briefing