A competition calling for research projects involving so-called AI scientists shows just how fast this technology is moving. A number of startups and universities that are building "AI scientists" to design and run experiments in the lab, including robot biologists and chemists, have just won extra funding from the UK government agency that funds moonshot R&D. The competition, set up by ARIA (the Advanced Research and Invention Agency), gives a clear sense of how fast this technology is moving: The agency received 245 proposals from research teams that are already building tools capable of automating increasing amounts of lab work. ARIA defines an AI scientist as a system that can run an entire scientific workflow, coming up with hypotheses, designing and running experiments to test those hypotheses, and then analyzing the results. In many cases, the system may then feed those results back into itself and run the loop again and again. Human scientists become overseers, coming up with the initial research questions and then letting the AI scientist get on with the grunt work.

While generative models for music composition are increasingly capable, their adoption by musicians is hindered by text-prompting, an asynchronous workflow disconnected from the embodied, responsive nature of instrumental performance. To address this, we introduce Aria-Duet, an interactive system facilitating a real-time musical duet between a human pianist and Aria, a state-of-the-art generative model, using a Yamaha Disklavier as a shared physical interface. The framework enables a turn-taking collaboration: the user performs, signals a handover, and the model generates a coherent continuation performed acoustically on the piano. Beyond describing the technical architecture enabling this low-latency interaction, we analyze the system's output from a musicological perspective, finding the model can maintain stylistic semantics and develop coherent phrasal ideas, demonstrating that such embodied systems can engage in musically sophisticated dialogue and open a promising new path for human-AI co-creation.

The rapid expansion of scientific data has widened the gap between analytical capability and research intent. Existing AI-based analysis tools, ranging from AutoML frameworks to agentic research assistants, either favor automation over transparency or depend on manual scripting that hinders scalability and reproducibility. We present ARIA (Automated Research Intelligence Assistant), a spec-driven, human-in-the-loop framework for automated and interpretable data analysis. ARIA integrates six interoperable layers, namely Command, Context, Code, Data, Orchestration, and AI Module, within a document-centric workflow that unifies human reasoning and machine execution. Through natural-language specifications, researchers define analytical goals while ARIA autonomously generates executable code, validates computations, and produces transparent documentation. Beyond achieving high predictive accuracy, ARIA can rapidly identify optimal feature sets and select suitable models, minimizing redundant tuning and repetitive experimentation. In the Boston Housing case, ARIA discovered 25 key features and determined XGBoost as the best performing model (R square = 0.93) with minimal overfitting. Evaluations across heterogeneous domains demonstrate ARIA's strong performance, interpretability, and efficiency compared with state-of-the-art systems. By combining AI for research and AI for science principles within a spec-driven architecture, ARIA establishes a new paradigm for transparent, collaborative, and reproducible scientific discovery.

Large language model (LLM) agents often struggle in environments where rules and required domain knowledge frequently change, such as regulatory compliance and user risk screening. Current approaches, like offline fine-tuning and standard prompting, are insufficient because they cannot effectively adapt to new knowledge during actual operation. To address this limitation, we propose the Adaptive Reflective Interactive Agent (ARIA), an LLM agent framework designed specifically to continuously learn updated domain knowledge at test time. ARIA assesses its own uncertainty through structured self-dialogue, proactively identifying knowledge gaps and requesting targeted explanations or corrections from human experts. It then systematically updates an internal, timestamped knowledge repository with provided human guidance, detecting and resolving conflicting or outdated knowledge through comparisons and clarification queries. We evaluate ARIA on the realistic customer due diligence name screening task on TikTok Pay, alongside publicly available dynamic knowledge tasks. Results demonstrate significant improvements in adaptability and accuracy compared to baselines using standard offline fine-tuning and existing self-improving agents. ARIA is deployed within TikTok Pay serving over 150 million monthly active users, confirming its practicality and effectiveness for operational use in rapidly evolving environments.

Precise 6-DoF simultaneous localization and mapping (SLAM) from onboard sensors is critical for wearable devices capturing egocentric data, which exhibits specific challenges, such as a wider diversity of motions and viewpoints, prevalent dynamic visual content, or long sessions affected by time-varying sensor calibration. While recent progress on SLAM has been swift, academic research is still driven by benchmarks that do not reflect these challenges or do not offer sufficiently accurate ground truth poses. In this paper, we introduce a new dataset and benchmark for visual-inertial SLAM with egocentric, multi-modal data. We record hours and kilometers of trajectories through a city center with glasses-like devices equipped with various sensors. We leverage surveying tools to obtain control points as indirect pose annotations that are metric, centimeter-accurate, and available at city scale. This makes it possible to evaluate extreme trajectories that involve walking at night or traveling in a vehicle. We show that state-of-the-art systems developed by academia are not robust to these challenges and we identify components that are responsible for this. In addition, we design tracks with different levels of difficulty to ease in-depth analysis and evaluation of less mature approaches. The dataset and benchmark are available at https://www.lamaria.ethz.ch.

You could be looking at the ultra-realistic and creepy future of sex dolls. A new'companion robot' made its debut at a major tech event in Vegas this week, sporting perfectly shaped breasts, a pert buttocks, thick lips and features of a youthful 20-something. Its creators, Realbotix, claim the 175,000 bot is to keep elderly gentlemen from being lonely. Asked why it had been shaped in an appealing way, the creators said it was because men wanted something nice to look at. Dressed in a black tracksuit, when asking the robot questions it gives long responses while having slightly jerky hand and body movements.

We introduce HOT3D, a publicly available dataset for egocentric hand and object tracking in 3D. The dataset offers over 833 minutes (more than 3.7M images) of multi-view RGB/monochrome image streams showing 19 subjects interacting with 33 diverse rigid objects, multi-modal signals such as eye gaze or scene point clouds, as well as comprehensive ground-truth annotations including 3D poses of objects, hands, and cameras, and 3D models of hands and objects. In addition to simple pick-up/observe/put-down actions, HOT3D contains scenarios resembling typical actions in a kitchen, office, and living room environment. The dataset is recorded by two head-mounted devices from Meta: Project Aria, a research prototype of light-weight AR/AI glasses, and Quest 3, a production VR headset sold in millions of units. Ground-truth poses were obtained by a professional motion-capture system using small optical markers attached to hands and objects. Hand annotations are provided in the UmeTrack and MANO formats and objects are represented by 3D meshes with PBR materials obtained by an in-house scanner. In our experiments, we demonstrate the effectiveness of multi-view egocentric data for three popular tasks: 3D hand tracking, 6DoF object pose estimation, and 3D lifting of unknown in-hand objects. The evaluated multi-view methods, whose benchmarking is uniquely enabled by HOT3D, significantly outperform their single-view counterparts.

In a cramped conference room in Bristol, Ilan Gur is trying to convince a group of plant biologists that they can change the world. The 44-year-old has the patter you'd expect from a Californian startup founder, but he's also one of the UK's most senior civil servants, so what comes next is unexpected. Close your eyes, he asks the scientists, and imagine pushing past the very edges of your research. The attendees take a beat, shifting slightly on their uncomfortable chairs. Positive visualization is not quite what they had expected from a workshop introducing them to the Advanced Research and Invention Agency (ARIA), the UK government's new high-risk, high-reward science funding agency.

The Advanced Research and Invention Agency (ARIA) will announce today that it's seeking proposals to work on systems for two related climate tipping points. One is the accelerating melting of the Greenland Ice Sheet, which could raise sea levels dramatically. The other is the weakening of the North Atlantic Subpolar Gyre, a huge current rotating counterclockwise south of Greenland that may have played a role in triggering the Little Ice Age around the 14th century. The goal of the five-year program will be to reduce scientific uncertainty about when these events could occur, how they would affect the planet and the species on it, and over what period those effects might develop and persist. In the end, ARIA hopes to deliver a proof of concept demonstrating that early warning systems can be "affordable, sustainable, and justified." No such dedicated system exists today, though there's considerable research being done to better understand the likelihood and consequences of surpassing these and other climate tipping points.

At the 2023 Defcon hacker conference in Las Vegas, prominent AI tech companies partnered with algorithmic integrity and transparency groups to sic thousands of attendees on generative AI platforms and find weaknesses in these critical systems. This "red-teaming" exercise, which also had support from the US government, took a step in opening these increasingly influential yet opaque systems to scrutiny. Now, the ethical AI and algorithmic assessment nonprofit Humane Intelligence is taking this model one step further. On Wednesday, the group announced a call for participation with the US National Institute of Standards and Technology, inviting any US resident to participate in the qualifying round of a nationwide red-teaming effort to evaluate AI office productivity software. The qualifier will take place online and is open to both developers and anyone in the general public as part of NIST's AI challenges, known as Assessing Risks and Impacts of AI, or ARIA.