PLAID is a multimodal generative model that simultaneously generates protein 1D sequence and 3D structure, by learning the latent space of protein folding models. What comes next after protein folding? In PLAID, we develop a method that learns to sample from the latent space of protein folding models to generate new proteins. It can accept compositional function and organism prompts, and can be trained on sequence databases, which are 2-4 orders of magnitude larger than structure databases. Unlike many previous protein structure generative models, PLAID addresses the multimodal co-generation problem setting: simultaneously generating both discrete sequence and continuous all-atom structural coordinates.

Managing your finances doesn't have to be a headache -- especially with the right budgeting app at your fingertips. Whether you're trying to track everyday spending, save for a big purchase or just keep a closer eye on your subscriptions, there's an app that can help. With Mint shutting down, plenty of users have been looking for the best budget apps to replace it, and luckily there are plenty of solid alternatives. From AI-powered spending trackers to apps that break down your expenses into easy-to-follow categories, the best budgeting tools help you take control of your money without the hassle of spreadsheets. Some focus on automating savings, while others give you a deep dive into your finances with powerful analytics and custom reporting. If you're still searching for the right Mint alternative, check out our guide to the best budgeting apps to replace Mint to find the best fit for your needs. If you're not sure where to start, we've rounded up the top budgeting apps to help you track spending, save smarter, and stick to your financial goals. No pun intended, but what I like about Quicken Simplifi is its simplicity. Whereas other budgeting apps try to distinguish themselves with dark themes and customizable emoji, Simplifi has a clean user interface, with a landing page that you just keep scrolling through to get a detailed overview of all your stats.

The PLAID (Performance-optimized Late Interaction Driver) algorithm for ColBERTv2 uses clustered term representations to retrieve and progressively prune documents for final (exact) document scoring. In this paper, we reproduce and fill in missing gaps from the original work. By studying the parameters PLAID introduces, we find that its Pareto frontier is formed of a careful balance among its three parameters; deviations beyond the suggested settings can substantially increase latency without necessarily improving its effectiveness. We then compare PLAID with an important baseline missing from the paper: re-ranking a lexical system. We find that applying ColBERTv2 as a re-ranker atop an initial pool of BM25 results provides better efficiency-effectiveness trade-offs in low-latency settings. However, re-ranking cannot reach peak effectiveness at higher latency settings due to limitations in recall of lexical matching and provides a poor approximation of an exhaustive ColBERTv2 search. We find that recently proposed modifications to re-ranking that pull in the neighbors of top-scoring documents overcome this limitation, providing a Pareto frontier across all operational points for ColBERTv2 when evaluated using a well-annotated dataset. Curious about why re-ranking methods are highly competitive with PLAID, we analyze the token representation clusters PLAID uses for retrieval and find that most clusters are predominantly aligned with a single token and vice versa. Given the competitive trade-offs that re-ranking baselines exhibit, this work highlights the importance of carefully selecting pertinent baselines when evaluating the efficiency of retrieval engines.

The HLTCOE team applied PLAID, an mT5 reranker, and document translation to the TREC 2023 NeuCLIR track. For PLAID we included a variety of models and training techniques -- the English model released with ColBERT v2, translate-train~(TT), Translate Distill~(TD) and multilingual translate-train~(MTT). TT trains a ColBERT model with English queries and passages automatically translated into the document language from the MS-MARCO v1 collection. This results in three cross-language models for the track, one per language. MTT creates a single model for all three document languages by combining the translations of MS-MARCO passages in all three languages into mixed-language batches. Thus the model learns about matching queries to passages simultaneously in all languages. Distillation uses scores from the mT5 model over non-English translated document pairs to learn how to score query-document pairs. The team submitted runs to all NeuCLIR tasks: the CLIR and MLIR news task as well as the technical documents task.

Tesla is reinventing its wheels. Its steering wheels, that is. The automaker's updated Model S and Model X unveiled on Wednesday feature all-new interior designs with rectangular steering wheels that look more like airplane yokes than what's typically found in a road car. They are also similar to what some racing cars use, not to mention the ultimate self-driving car: K.I.T.T. from the "Knight Rider" TV show. Tesla had previously incorporated into the design of its Cybertruck and Roadster prototypes.

Byron Lichtenstein is a principal at Insight Partners, and sees most opportunity in the convergence between software and payments. Insight Partners focuses mainly on growth-stage software companies across verticals from education to social media to fintech, and it has invested in German neobank N26, business expense management startup Divvy, and payment fraud monitoring startup Sift. Here are the ways he sees payments and software coming together to find value in a changing industry. "Historically, we've always been software investors," Lichtenstein told Business Insider. "What's changed over the past two years that we found really interesting is that dumb payments don't really --obviously, they exist --but they're not really a thing anymore," Lichtenstein said.

Deep reinforcement learning has demonstrated increasing capabilities for continuous control problems, including agents that can move with skill and agility through their environment. An open problem in this setting is that of developing good strategies for integrating or merging policies for multiple skills, where each individual skill is a specialist in a specific skill and its associated state distribution. We extend policy distillation methods to the continuous action setting and leverage this technique to combine expert policies, as evaluated in the domain of simulated bipedal locomotion across different classes of terrain. We also introduce an input injection method for augmenting an existing policy network to exploit new input features. Lastly, our method uses transfer learning to assist in the efficient acquisition of new skills. The combination of these methods allows a policy to be incrementally augmented with new skills. We compare our progressive learning and integration via distillation (PLAID) method against three alternative baselines.

For example, the algorithm learned to recognize whether people were wearing a jacket, a scarf, a necktie, glasses, a hat, and so on. The clustering algorithm found some 400 different visual themes, such as people wearing white T-shirts and glasses, or wearing red V-neck tops or black dresses, or not wearing tops at all! They also found a sudden increase in popularity of yellow shirts in Colombia and Brazil during the June/July 2014 football World Cup--both countries' football teams wear yellow. As Matzen and co conclude: "The combination of big data, machine learning, computer vision, and automated analysis algorithms would make for a very powerful analysis tool more broadly in visual discovery of fashion and many other areas."