Eggie, Neo, Isaac and Memo are domestic robots. But would you let them load your dishwasher? The idea of having a friendly robot butler that can do all the dull duties of running a home has existed for decades. But now, thanks to AI, it's genuinely happening and this year the first truly multi-purpose domestic bots will start to enter homes. In Silicon Valley, they're being trained at speed to fold laundry, load the dishwasher, and clean up after us.

Test-Time Adaptation (TT A) methods are often computationally expensive, require a large amount of data for effective adaptation, or are brittle to hyperparam-eters. Based on a theoretical foundation of the geometry of the latent space, we are able to significantly improve the alignment between source and distribution-shifted samples by re-centering target data embeddings at the origin. This insight motivates NEO - a hyperparameter-free fully TT A method, that adds no significant compute compared to vanilla inference. NEO is able to improve the classification accuracy of ViT -Base on ImageNet-C from 55.6% to 59.2% after adapting on just one batch of 64 samples. When adapting on 512 samples NEO beats all 7 TT A methods we compare against on ImageNet-C, ImageNet-R and ImageNet-S and beats 6/7 on CIFAR-10-C, while using the least amount of compute. NEO performs well on model calibration metrics and additionally is able to adapt from 1 class to improve accuracy on 999 other classes in ImageNet-C. On Raspberry Pi and Jetson Orin Nano devices, NEO reduces inference time by 63% and memory usage by 9% compared to baselines. Our results based on 3 ViT architectures and 4 datasets show that NEO can be used efficiently and effectively for TT A. A central challenge in machine learning is maintaining performance under distribution shifts between training and deployment. For instance, an image classifier may excel on curated training data but degrade on real-world inputs with snow, fog, or motion blur. Test-Time Adaptation (TT A) methods (Li et al., 2018; Wang et al., 2024; Liang et al., 2020; Wang et al., 2021; Niu et al., 2023) address this by leveraging unlabeled test samples without requiring access to training data, making them particularly suited to the modern setting of large pre-trained models.

Autonomous navigation by drones using onboard sensors, combined with deep learning and computer vision algorithms, is impacting a number of domains. We examine the use of drones to autonomously follow and assist Visually Impaired People (VIPs) in navigating urban environments. Estimating the absolute distance between the drone and the VIP, and to nearby objects, is essential to design obstacle avoidance algorithms. Here, we present NeoARCADE (Neo), which uses depth maps over monocular video feeds, common in consumer drones, to estimate absolute distances to the VIP and obstacles. Neo proposes robust calibration technique based on depth score normalization and coefficient estimations to translate relative distances from depth map to absolute ones. It further develops a dynamic recalibration method that can adapt to changing scenarios. We also develop two baseline models, Regression and Geometric, and compare Neo with SOTA depth map approaches and the baselines. We provide detailed evaluations to validate their robustness and generalizability for distance estimation to VIPs and other obstacles in diverse and dynamic conditions, using datasets collected in a campus environment. Neo predicts distances to VIP with an error <30cm, and to different obstacles like cars and bicycles within a maximum error of 60cm, which are better than the baselines. Neo also clearly out-performs SOTA depth map methods, reporting errors up to 5.3-14.6x lower.

Large-language-model (LLM) agents exhibit complex, context-sensitive behaviour that quickly renders static benchmarks and ad-hoc manual testing obsolete. We present Neo, a configurable, multi-agent framework that automates realistic, multi-turn evaluation of LLM-based systems. Neo couples a Question Generation Agent and an Evaluation Agent through a shared context-hub, allowing domain prompts, scenario controls and dynamic feedback to be composed modularly. Test inputs are sampled from a probabilistic state model spanning dialogue flow, user intent and emotional tone, enabling diverse, human-like conversations that adapt after every turn. Applied to a production-grade Seller Financial Assistant chatbot, Neo (i) uncovered edge-case failures across five attack categories with a 3.3% break rate close to the 5.8% achieved by expert human red-teamers, and (ii) delivered 10-12X higher throughput, generating 180 coherent test questions in around 45 mins versus 16h of human effort. Beyond security probing, Neo's stochastic policies balanced topic coverage and conversational depth, yielding broader behavioural exploration than manually crafted scripts. Neo therefore lays a foundation for scalable, self-evolving LLM QA: its agent interfaces, state controller and feedback loops are model-agnostic and extensible to richer factual-grounding and policy-compliance checks. We release the framework to facilitate reproducible, high-fidelity testing of emerging agentic systems.

We present a comprehensive analysis of the digest2 parameters for candidates of the Near-Earth Object Confirmation Page (NEOCP) that were reported between 2019 and 2024. Our study proposes methods for significantly reducing the inclusion of non-NEO objects on the NEOCP. Despite the substantial increase in near-Earth object (NEO) discoveries in recent years, only about half of the NEOCP candidates are ultimately confirmed as NEOs. Therefore, much observing time is spent following up on non-NEOs. Furthermore, approximately 11% of the candidates remain unconfirmed because the follow-up observations are insufficient. These are nearly 600 cases per year. To reduce false positives and minimize wasted resources on non-NEOs, we refine the posting criteria for NEOCP based on a detailed analysis of all digest2 scores. We investigated 30 distinct digest2 parameter categories for candidates that were confirmed as NEOs and non-NEOs. From this analysis, we derived a filtering mechanism based on selected digest2 parameters that were able to exclude 20% of the non-NEOs from the NEOCP while maintaining a minimal loss of true NEOs. We also investigated the application of four machine-learning (ML) techniques, that is, the gradient-boosting machine (GBM), the random forest (RF) classifier, the stochastic gradient descent (SGD) classifier, and neural networks (NN) to classify NEOCP candidates as NEOs or non-NEOs. Based on digest2 parameters as input, our ML models achieved a precision of approximately 95% in distinguishing between NEOs and non-NEOs. Results. Combining the digest2 parameter filter with an ML-based classification model, we demonstrate a significant reduction in non-NEOs on the NEOCP that exceeds 80%, while limiting the loss of NEO discovery tracklets to 5.5%. Importantly, we show that most follow-up tracklets of initially misclassified NEOs are later correctly identified as NEOs.

After creating a stir with the 200 Neo, DJI is back at it with another innovative drone, the Flip. It has a first-of-a-kind folding design and shrouded propellers to keep people safe. It also integrates 3D infrared obstacle detection to track subjects and has a long list of impressive features. With a camera borrowed from the Mini 4 Pro, the Flip can take high-quality 4K 60p video indoors or out with little risk. It comes with vlogger-friendly features like Direction Track and Quickshots for social media.

DJI continues its streak of innovative (and highly leaked) drones with the launch of the Flip, a lightweight and people-safe model that folds in a new direction -- downward -- to accommodate the large, shrouded propellers. The new model should appeal to beginners and experienced users alike with features like a large sensor, 4K 100p video, safety features, a three-axis gimbal and an affordable price. The company says the Flip "combine[s] the simplicity of the DJI Neo with the stunning photo capabilities of the DJI Mini," but in many ways, it's better than both. It borrows a LiDAR system from the Air 3S for obstacle detection and the Flip's propellers are protected on all sides, making it all but impossible to hurt someone with them. DJI says the support structure for the guards is made of carbon fiber string that's 1/60th the weight of polycarbonate material and just as strong.

Aside from the global shutter on Sony's A9 III and some cool mirrorless options -- the Fujifilm X100 VI, Panasonic S9 and Canon EOS R5 II come to mind -- 2024 was a dull year for cameras full of small tweaks and minor improvements. For 200, aerial photography is now finally in reach for just about anyone. DJI released its product lineup this year with a sword of Damocles hanging over its head: the US government was planning to ban sales of the company's products by the end of 2024 over potential fears of spying. It was only at the last minute that DJI gained a reprieve, thanks in large part to lobbying by public safety groups that heavily rely on its drones. It now has until the end of 2025 to prove that its products don't pose a risk.

When DJI revealed its tiny 200 Neo drone, I immediately saw how it could fit into my vlogger's toolkit to supplement my Mini 4 Pro and Mavic 3 Pro. Flying those sophisticated drones is a whole thing that requires planning. But the Neo can be launched spontaneously to grab quick and fun shots, thanks to features like palm takeoff and voice control. That ease of use also makes it ideal for the social media influencers. You get features from DJI's bigger drones like ActiveTrack, FPV capabilities and even support for DJI's Mic 2. And forget about the fuzzy video you may have seen on other cheap drones. The Neo can record in sharp 4K, making it suitable for content creators who need affordable aerial video.