Google has been using AI in its shopping tools for a bit now, letting people use generative AI to " try on" clothes and see if the garments look good on them, or look for things using Google Lens. Today, Google is implementing even more AI functionality in its Shopping service, allowing Gemini to "show the most relevant products." Instead of only showing an assortment of products, Google Shopping now includes an AI-generated brief that recommends other products associated with what you searched for. For example, shoppers looking for notebooks may get a brief mentioning stationery like pens and erasers. The products shown are also sourced from sources like articles and guides from across the web.

The fashion industry is one of the leading domains in the global e-commerce sector, prompting major online retailers to employ recommendation systems for product suggestions and customer convenience. While recommendation systems have been widely studied, most are designed for general e-commerce problems and struggle with the unique challenges of the fashion domain. To address these issues, we propose a sequential fashion recommendation framework that leverages a pre-trained large language model (LLM) enhanced with recommendation-specific prompts. Our framework employs parameter-efficient fine-tuning with extensive fashion data and introduces a novel mix-up-based retrieval technique for translating text into relevant product suggestions. Extensive experiments show our proposed framework significantly enhances fashion recommendation performance.

Online Kernel Learning (OKL) has attracted considerable research interest due to its promising predictive performance in streaming environments. Second-order approaches are particularly appealing for OKL as they often offer substantial improvements in regret guarantees. However, existing second-order OKL approaches suffer from at least quadratic time complexity with respect to the pre-set budget, rendering them unsuitable for meeting the real-time demands of large-scale streaming recommender systems. The singular value decomposition required to obtain explicit feature mapping is also computationally expensive due to the complete decomposition process. Moreover, the absence of incremental updates to manage approximate kernel space causes these algorithms to perform poorly in adversarial environments and real-world streaming recommendation datasets. To address these issues, we propose FORKS, a fast incremental matrix sketching and decomposition approach tailored for second-order OKL. FORKS constructs an incremental maintenance paradigm for second-order kernelized gradient descent, which includes incremental matrix sketching for kernel approximation and incremental matrix decomposition for explicit feature mapping construction. Theoretical analysis demonstrates that FORKS achieves a logarithmic regret guarantee on par with other second-order approaches while maintaining a linear time complexity w.r.t. the budget, significantly enhancing efficiency over existing approaches. We validate the performance of FORKS through extensive experiments conducted on real-world streaming recommendation datasets, demonstrating its superior scalability and robustness against adversarial attacks.

Object navigation in unknown environments is crucial for deploying embodied agents in real-world applications. While we have witnessed huge progress due to large-scale scene datasets, faster simulators, and stronger models, previous studies mainly focus on limited scene types and target objects. In this paper, we study a new task of navigating to diverse target objects in a large number of scene types. To benchmark the problem, we present a large-scale scene dataset, DivScene, which contains 4,614 scenes across 81 different types. With the dataset, we build an end-to-end embodied agent, NatVLM, by fine-tuning a Large Vision Language Model (LVLM) through imitation learning. The LVLM is trained to take previous observations from the environment and generate the next actions. We also introduce CoT explanation traces of the action prediction for better performance when tuning LVLMs. Our extensive experiments find that we can build a performant LVLM-based agent through imitation learning on the shortest paths constructed by a BFS planner without any human supervision. Our agent achieves a success rate that surpasses GPT-4o by over 20%. Meanwhile, we carry out various analyses showing the generalization ability of our agent. Our code and data are available at https://github.com/zhaowei-wang-nlp/DivScene.

Knowledge graphs constantly evolve with new entities emerging, existing definitions being revised, and entity relationships changing. These changes lead to temporal degradation in entity linking models, characterized as a decline in model performance over time. To address this issue, we propose leveraging graph relationships to aggregate information from neighboring entities across different time periods. This approach enhances the ability to distinguish similar entities over time, thereby minimizing the impact of temporal degradation. We introduce \textbf{CYCLE}: \textbf{C}ross-\textbf{Y}ear \textbf{C}ontrastive \textbf{L}earning for \textbf{E}ntity-Linking. This model employs a novel graph contrastive learning method to tackle temporal performance degradation in entity linking tasks. Our contrastive learning method treats newly added graph relationships as \textit{positive} samples and newly removed ones as \textit{negative} samples. This approach helps our model effectively prevent temporal degradation, achieving a 13.90\% performance improvement over the state-of-the-art from 2023 when the time gap is one year, and a 17.79\% improvement as the gap expands to three years. Further analysis shows that CYCLE is particularly robust for low-degree entities, which are less resistant to temporal degradation due to their sparse connectivity, making them particularly suitable for our method. The code and data are made available at \url{https://github.com/pengyu-zhang/CYCLE-Cross-Year-Contrastive-Learning-in-Entity-Linking}.

Amazon's October Prime Day (aka Prime Big Deal Days) officially ended Oct. 9, but that doesn't mean every deal is done. Some companies have extended the discounts on their gear a little longer, and many no longer require you to have an active Amazon Prime subscription to take advantage of them. Some are still Prime subscriber exclusive, though, and if that's the case, you can sign up for a free 30-day trial here if you don't have one. However, you should add to cart now; there's no telling how long any of these discounts will last since they're no longer tied to an official sale. Google's Nest Thermostat is a particularly good Prime Day deal because it can actually help pay for itself over time. The Wi-Fi-enabled thermostat can learn your habits and schedule over time and activate your HVAC (Heating, Ventilation, and Air Conditioning) system more efficiently and less frequently.

There are 45 Prime Day tech deals under 50 that are still live for the second day of Amazon's sale. Here, we've gathered up the smaller gadgets and supporting accessories, brands and devices we've tested and reviewed and know to be worth your money. This list includes useful gear like chargers, storage cards, cables, batteries and even earbuds -- plus a few Lego sets thrown in for good measure. Here are the best Prime Day Tech deals under 50. If you've got 25 and some change in an account somewhere, you can get something decent from Amazon's sale (particularly if you're a Prime member and don't have to pay for shipping).

Nobody wants to sweep or mop their floors, and with Eureka's J20, you don't have to. This smart-home vacuum, which allows you to monitor and schedule cleanings using Wi-Fi, is a massive improvement over previous generations of similar hardware. The robot vacuum is also currently 170 cheaper than usual thanks to a deal during Amazon's October Prime Day (aka Prime Big Deal Days). Remember, if you don't have an active Amazon Prime subscription, you can sign up for a free 30-day trial here, which is the perfect price to save you time, trouble, and almost 200. The premium smart-home cleaning device has every feature you'd need, from a maximum suction of 8000Pa (pascals), the ability to clean both hardwood and carpeted surfaces, and, critically, the intelligence to return to its dock to recharge and clean its brushes. Robot vacuums have become commoditized over the past few years, but the J20 stands out due to its raw performance.

Amazon said on Wednesday that it's rolling out new online ordering methods for Prime members, including the ability to bundle standard orders and groceries in one same-day shipment. The company is also adding more combined Amazon / Whole Foods fulfillment centers and trialing a store where robots pack your Amazon orders while you shop for groceries. Customers there can shop "tens of thousands of grocery items" (including fresh ones) alongside regular Amazon orders for things like AirPods or Lego sets. The items will be bundled in one order and arrive together in a user-selected, same-day or overnight delivery window. The company plans to expand the combined same-day model to more areas after it tests and learns from the Phoenix trial.

While robot vacuums are expensive, they're a great way to automate a chore that needs to get done regularly. If you're looking to make your home a bit smarter, or you simply hate vacuuming, October Prime Day (and sale events like it) can make it easier to jump into the world of robo-vacs by discounting top models by hundreds of dollars. October Prime Day has brought some of the lowest prices we've seen to many of our top picks for the best robot vacuums available today, including models from iRobot and Shark. There are even some solid cordless stick vacuums on sale as well as part of the Prime Day deals. We collected all of the best Prime Day deals on vacuums here so you don't have to go searching for them.