Large language models (LLMs) often need to incorporate external knowledge to solve theme-specific problems. Retrieval-augmented generation (RAG) has shown its high promise, empowering LLMs to generate more qualified responses with retrieved external data and knowledge. However, most RAG methods retrieve relevant documents based on either sparse or dense retrieval methods or their combinations, which overlooks the essential, multi-dimensional, and structured semantic information present in documents. This structured information plays a critical role in finding concise yet highly relevant information for domain knowledge-intensive tasks, such as scientific question-answering (QA). In this work, we introduce a multi-dimensional (cube) structure, Hypercube, which can index and allocate documents in a pre-defined multi-dimensional space. Built on the hypercube, we further propose Hypercube-RAG, a novel RAG framework for precise and efficient retrieval. Given a query, Hypercube-RAG first decomposes it based on its entities, phrases, and topics along with pre-defined hypercube dimensions, and then retrieves relevant documents from cubes by aligning these decomposed components with corresponding dimensions. Experiments on three datasets across different domains demonstrate that our method improves response accuracy by 3.7% and retrieval accuracy by 5.3% over the strongest RAG baseline. It also boosts retrieval efficiency (speed) by one or two magnitudes faster than graph-based RAG. Notably, our Hypercube-RAG inherently offers explainability by revealing those underlying dimensions used for retrieval. The code and data are available at https://github.com/JimengShi/Hypercube-RAG.

An experienced human Observer reading a document -- such as a crime report -- creates a succinct plot-like $\textit{``Working Memory''}$ comprising different actors, their prototypical roles and states at any point, their evolution over time based on their interactions, and even a map of missing Semantic parts anticipating them in the future. $\textit{An equivalent AI Observer currently does not exist}$. We introduce the $\textbf{[G]}$enerative $\textbf{[S]}$emantic $\textbf{[W]}$orkspace (GSW) -- comprising an $\textit{``Operator''}$ and a $\textit{``Reconciler''}$ -- that leverages advancements in LLMs to create a generative-style Semantic framework, as opposed to a traditionally predefined set of lexicon labels. Given a text segment $C_n$ that describes an ongoing situation, the $\textit{Operator}$ instantiates actor-centric Semantic maps (termed ``Workspace instance'' $\mathcal{W}_n$). The $\textit{Reconciler}$ resolves differences between $\mathcal{W}_n$ and a ``Working memory'' $\mathcal{M}_n^*$ to generate the updated $\mathcal{M}_{n+1}^*$. GSW outperforms well-known baselines on several tasks ($\sim 94\%$ vs. FST, GLEN, BertSRL - multi-sentence Semantics extraction, $\sim 15\%$ vs. NLI-BERT, $\sim 35\%$ vs. QA). By mirroring the real Observer, GSW provides the first step towards Spatial Computing assistants capable of understanding individual intentions and predicting future behavior.

The objective of this paper is to employ machine learning (ML) and deep learning (DL) techniques to obtain from input data (storm features) available in or derived from the HURDAT2 database models capable of simulating important hurricane properties such as landfall location and wind speed that are consistent with historical records. In pursuit of this objective, a trajectory model providing the storm center in terms of longitude and latitude, and intensity models providing the central pressure and maximum 1-$min$ wind speed at 10 $m$ elevation were created. The trajectory and intensity models are coupled and must be advanced together, six hours at a time, as the features that serve as inputs to the models at any given step depend on predictions at the previous time steps. Once a synthetic storm database is generated, properties of interest, such as the frequencies of large wind speeds may be extracted from any part of the simulation domain. The coupling of the trajectory and intensity models obviates the need for an intensity decay inland of the coastline. Prediction results are compared to historical data, and the efficacy of the storm simulation models is demonstrated for three examples: New Orleans, Miami and Cape Hatteras.

For me everything is evolution," he says. Nor is he a big fan of the "gig economy", the use of often poorly paid contractors, which has been used to power so many other "disruptive" tech platforms like Uber and Lyft. Reef kitchen employees are staff and get paid $20 an hour. "Our approach from the beginning was that this has to be a business built on ownership and accountability," he says. "In general I don't believe in the fundamentals of building a business on gig workers.

A glossy high rise in the heart of Miami aims to be the first residential building in the U.S. with a specially designed rooftop to accommodate a Jetsons-like future where cars take to the skies. Halfway through the construction of Paramount Miami World Center, developers determined that the $4 billion, 60-story complex needed something extra to stand out among the vast array of living options for the super-rich. So they installed an observation deck at the top that doubles as a landing pad for vertical takeoff and landing vehicles, often called VTOLs, or flying cars. The tower will have its grand opening early in 2020. Meanwhile, a flying car's reality, where passengers can be dropped off at home like Amazon drone packages, could be decades away – if ever.

Anyone summoning a Ford self-driving taxi after the company's service launches in 2021 could be offered detours to sponsoring stores or wind up riding alongside packages out for delivery. At an event on Wednesday to showcase Ford's progress in developing autonomous vehicles, the carmaker said its driverless rides could be less than half the price of today's ride-share journeys,if the cars were used day and night and carried interactive adverts. The firm is also insistent that it is not lagging behind Waymo, which has promised a commercial self-driving taxi service by the end of this year, or GM, which says it will follow suit in 2019. "If we wanted to call a launch 100 vehicles [on the road] next year and go into some business, we could do that," said Sherif Marakby, CEO of Ford Autonomous Vehicles, a spin-off from the giant automaker tasked with developing autonomous technologies, vehicles, and services. "[But] we're an auto company and when we talk about launch at scale, we're talking tens of thousands vehicles, and [doing that] profitably. That's different from what others are thinking."

Welcome to the 21st century, where a request for extra towels in your hotel room may be answered by a roughly 4-foot-high purple robot on wheels. Miami YotelPad -- an unfinished 30-story mixed-use development in downtown Miami -- will employ three robot butlers for its guests. These robots don't look like humans (thankfully?), but they're programmed to execute tasks normally left to their biological counterparts: delivering room service, playing music and even engaging in conversation. No one covers what is happening in our community better than we do. And with a digital subscription, you'll never miss a local story.

Ford is venturing well beyond its home turf of Michigan to test its fledgling self-driving delivery service. The automaker has revealed a pilot program that will see its autonomous vehicles roaming the streets of downtown Miami and Miami Beach. The initial test will separate the delivery and self-driving products, and will gauge what works for both customers and companies. How do you pick up your delivery from a self-driving car, for example, and how far are people willing to walk to get their grub? Domino's is active right now, while Postmates should be available in March.

If you work on self-driving cars, the cocktail party question people always ask is probably: When will I get to interact with one? For two years now, Ford Motor Company has had an answer--in 2021. That year, Ford wants to launch a self-driving taxi service, and it wants to start making deliveries with driverless vehicles. But before the Detroit automaker does any of that, it needs to figure out how to run a fleet. Which is why the company announced today that it will begin to test autonomous vehicles and build its first operations terminal in Miami, Florida.

Donald Trump invoked the possibility of a violent assault on Hillary Clinton once again on Saturday, a day after he suggested that her Secret Service bodyguards disarm and "let's see what happens." In a post Saturday morning on Twitter, Trump falsely accused Clinton of trying to take away Americans' 2nd Amendment rights, just as he did Friday night at a Miami rally where he said her Secret Service agents should "drop all weapons." "Will guns be taken from her heavily armed Secret Service detail? Trump said Friday night that Clinton's Secret Service detail should disarm because she supports gun control. "What do you think, yes?" he asked the crowd. Let's see what happens to her. Take their guns away, OK? It would be very dangerous."