Open Problems in Frontier AI Risk Management
arXiv:2604.25982v1 Announce Type: cross Abstract: Frontier AI both amplifies existing risks and introduces qualitatively novel challenges. Not only is there a notable lack of stable scientific consens
Knowledge catalogue
arXiv:2604.25982v1 Announce Type: cross Abstract: Frontier AI both amplifies existing risks and introduces qualitatively novel challenges. Not only is there a notable lack of stable scientific consens
OpenAI: OpenAI says its models, starting with GPT-5.1, “increasingly mentioned goblins, gremlins, and other creatures”, leading to prompt instructions to mitigate it — Starting with GPT-5.1, our model
Families of victims of a February 2026 mass shooting in Tumbler Ridge, British Columbia sued OpenAI and CEO Sam Altman, alleging the company identified the shooter as a credible threat eight months be
OpenAI is opening up about its goblin problem. After a report from Wired revealed instructions to OpenAI's coding model to 'never talk about goblins, gremlins, raccoons, trolls, ogres, pigeons, or oth
OpenAI’s lawyer plays dirty. Why I am not surprised? 🚨 OpenAI's lawyer read a 2016 email from Sutskever to Musk to argue Musk knew OpenAI would go closed-source Sutskever: 'As we get closer to buildin
OpenAI is preparing to launch a new frontier cybersecurity model, GPT-5.5-Cyber. CEO Sam Altman said the model will not be available to the general public, but will be first rolled out to a select gro
Semantic intelligence company Collate Inc. today announced the launch of Collate AI Analytics, a new chat-based tool that lets data analysts find data sources, write queries and build dashboards from
arXiv:2604.26091v1 Announce Type: new Abstract: We study reliability in autonomous language-model agents that translate user mandates into validated tool actions under real capital. The setting is DX
arXiv:2507.17544v4 Announce Type: replace-cross Abstract: Differential privacy has become a cornerstone in the development of privacy-preserving learning algorithms. This work addresses optimizing dif
arXiv:2604.26206v1 Announce Type: cross Abstract: A predecessor pilot (Cacioli, 2026) found that Llama-3-8B implements prompted sandbagging as positional collapse rather than answer avoidance. However
Oracle has dropped 50% since I declared that the OpenAI deal they made was “peak bubble”. And, as @edzitron lays out below, they may fall a lot further yet. At which point the entire house of cards co
arXiv:2604.26472v1 Announce Type: cross Abstract: We study sequential interventions under prerequisite constraints. In this setting, admissible intervention sequences are paths in the ideal lattice of
arXiv:2505.11669v3 Announce Type: replace-cross Abstract: We address the computational and theoretical limitations of current distributional alignment methods for source-free unsupervised domain adapt
Ouch! Current AI assistants often corrupt documents. Sounds like an intern you can’t trust — once again. A trillion dollar investment in scaling hasn’t solved this. New Microsoft paper shows that curr
Our agent harness makes models inside Cursor faster, smarter, and more token-efficient. Here's how we test improvements to the harness, monitor and repair degradations, and customize it for different
Our evaluation of OpenAI's GPT-5.5 cyber capabilities The UK's AI Security Institute previously evaluated Claude Mythos: now they've evaluated GPT-5.5 for finding security vulnerability and found it t
arXiv:2505.14808v2 Announce Type: replace-cross Abstract: The transformer's remarkable ability to perform in-context learning (ICL) has sparked a wide range of studies designed to understand its stren
I cannot provide a summary for this entry as the title appears incomplete or corrupted, making it impossible to determine the actual content or claims being discussed. The URL format suggests this is
arXiv:2603.05959v3 Announce Type: replace Abstract: Reconstructing 3D geometry from streaming video requires continuous inference under bounded resources. Recent geometric foundation models achieve im
arXiv:2604.25602v2 Announce Type: replace Abstract: Deploying production-ready multi-agent systems (MAS) in complex industrial environments remains challenging due to limitations in scalability, obser
p5.js is a JavaScript library for creative coding that enables artists and designers to create visual art, animations, and interactive experiences through code. This Nous Research skill module likely
arXiv:2604.26573v1 Announce Type: new Abstract: Improving large language model (LLM) reasoning requires supervision that is both aligned with the model's own test-time states and informative at the to
arXiv:2604.26675v1 Announce Type: cross Abstract: We investigate variational quantum classifiers (VQCs) for land-cover classification from multispectral satellite imagery, adopting a feature-map persp
arXiv:2509.23410v4 Announce Type: replace-cross Abstract: Large language models (LLMs) deliver impressive performance but incur prohibitive memory and compute costs at deployment. Model pruning is an
arXiv:2507.19067v2 Announce Type: replace-cross Abstract: Recommender systems based on graph neural networks (GNNs) have been proved to perform well on user-item interactions. However, they commonly s
PEOPLE ARE NOW RUNNING CLAUDE CODE WITH LOCAL AI MODELS TO AVOID API COSTS. By connecting tools like Ollama and Gemma 4, developers can build apps locally with unlimited usage and no monthly billing.
arXiv:2601.06287v2 Announce Type: replace Abstract: The Third Perception Test challenge was organised as a full-day workshop alongside the IEEE/CVF International Conference on Computer Vision (ICCV) 2
arXiv:2604.26527v1 Announce Type: cross Abstract: Human-robot interaction is emerging as an important paradigm for integrating persons with disabilities into the workplace. While these systems can ena
arXiv:2604.26233v1 Announce Type: new Abstract: As Large Language Models (LLMs) are proposed as legal decision assistants, and even first-instance decision-makers, across a range of judicial and admin
Pi + @huggingface llamacpp + Qwen3.6 = 🔥🔥🔥🔥 turns out not killing the prefix cache all the time and notnhaving a humongous set of tools and a massive system prompt is good for local model use. who'd h
This post compares Pi and Qwen 3.6 AI models to Tesla's Full Self-Driving (FSD) technology, suggesting that advanced local AI models enable autonomous computing capabilities similar to self-driving ve
arXiv:2604.26593v1 Announce Type: new Abstract: Digital twins provide a powerful paradigm for diagnostic and prognostic tasks in the monitoring and control of engineered systems; however, their deploy
arXiv:2604.25945v1 Announce Type: cross Abstract: Wireless radiance field (WRF) reconstruction aims to learn a continuous, queryable representation of radio frequency characteristics over 3D space and
arXiv:2604.26318v1 Announce Type: new Abstract: Point cloud registration (PCR) is a fundamental task for integrating 3D observations in remote sensing applications. This paper proposes a fast and effe
arXiv:2603.04337v2 Announce Type: replace-cross Abstract: Constructing computer-aided design (CAD) models is labor-intensive but essential for engineering and manufacturing. Recent advances in Large L
arXiv:2604.26078v1 Announce Type: new Abstract: Photoplethysmography (PPG) is increasingly used in wearable affective computing due to its low cost and ease of integration into consumer devices. Recen
This article emphasizes that healthcare organizations must move beyond simply predicting hospital readmissions to taking timely preventive actions based on those predictions. It likely discusses how m
arXiv:2604.26561v1 Announce Type: cross Abstract: Multi-agent deliberation systems using large language models (LLMs) are increasingly proposed for policy simulation, yet they suffer from artificial c
arXiv:2603.03664v2 Announce Type: replace-cross Abstract: Learning-to-communicate (LTC) in partially observable environments has received increasing attention in deep multi-agent reinforcement learnin
arXiv:2604.26184v1 Announce Type: new Abstract: A privacy-preserving clothing classification scheme is presented to enable secure occupant-centric control (OCC) systems. Although the utilization of ca
arXiv:2604.26073v1 Announce Type: cross Abstract: Industrial chemical plants often operate under strict data confidentiality constraints, making centralized data-driven process modeling difficult. Fed
arXiv:2604.26366v1 Announce Type: cross Abstract: Data quality assessment is an essential step that ensures the reliability of the subsequent structural health monitoring (SHM) tasks. This study propo
arXiv:2603.15262v2 Announce Type: replace Abstract: Modern e-commerce search is evolving to resolve complex user intents. While Large Language Models (LLMs) offer strong reasoning, existing LLM-based
arXiv:2604.26943v1 Announce Type: new Abstract: We introduce ProcFunc, a library for Blender-based procedural 3D generation in Python. ProcFunc provides a library of easy-to-use Python functions, whic
arXiv:2604.26508v1 Announce Type: cross Abstract: Deploying Vision-Language Models (VLMs) on edge devices remains challenging due to their substantial computational and memory demands, which exceed th
This post was written in April 2026. Cloud products, feature maturity, and recommended patterns change over time, so readers should treat these examples as directional guidance. For teams already usin
arXiv:2604.06061v2 Announce Type: replace Abstract: Text-to-image generation has progressed rapidly, but faithfully generating complex scenes requires extensive trial-and-error to find the exact promp
arXiv:2604.17612v2 Announce Type: replace-cross Abstract: Multi-agent systems built on large language models (LLMs) are difficult to reason about. Coordination errors such as deadlocks or type-mismatc
arXiv:2512.23726v2 Announce Type: replace-cross Abstract: The 3D fast silent multi-parametric mapping sequence with zero echo time (MuPa-ZTE) is a novel quantitative MRI (qMRI) acquisition that enable
arXiv:2604.26018v1 Announce Type: cross Abstract: We introduce QERNEL, a foundational neural wavefunction that variationally solves families of parameterized many-electron Hamiltonians and captures th
arXiv:2604.26505v1 Announce Type: cross Abstract: Dynamic quantization emerged as a practical approach to increase the utilization and efficiency of the machine learning serving flow. Unlike static qu
arXiv:2412.11399v4 Announce Type: replace Abstract: Driven by global climate change and the ongoing energy transition, the coupling between power supply capabilities and meteorological factors has bec
arXiv:2604.26834v1 Announce Type: cross Abstract: We present a quantum feature-selection framework based on a higher-order unconstrained binary optimization (HUBO) formulation that explicitly incorpor
arXiv:2604.26413v1 Announce Type: cross Abstract: This paper presents Quantum Gatekeeper, a context-bound image steganography framework where successful payload recovery depends on both cryptographic
It's a common misconception that we can't tell who is using LLM and who is not. I'm sure we didn't catch 100% of LLM-assisted PRs over the past few months, but the kind of mistakes humans make are fun
arXiv:2604.26213v1 Announce Type: cross Abstract: Loading high dimensional distributions is an important task for utilizing quantum computers on applications ranging from machine learning to finance.
arXiv:2604.26435v1 Announce Type: cross Abstract: The rapid advancement of object detection architectures has positioned single stage detectors as the dominant solution for real-time visual perception
Variational Joint Embedding (VJE) is a framework that synthesizes joint embedding and variational inference to enable self-supervised learning of probabilistic representations in a reconstruction-free
arXiv:2510.08547v2 Announce Type: replace-cross Abstract: Towards the aim of generalized robotic manipulation, spatial generalization is the most fundamental capability that requires the policy to wor
arXiv:2604.26067v1 Announce Type: new Abstract: We present RADIO-ViPE (Reduce All Domains Into One -- Video Pose Engine), an online semantic SLAM system that enables geometry-aware open-vocabulary gro