Different entities with the same name can be difficult to distinguish. Handling confusing entity mentions is a crucial skill for language models (LMs). For example, given the question "Where was Michael Jordan educated?" and a set of documents discussing different people named Michael Jordan, can LMs distinguish entity mentions to generate a cohesive answer to the question? To test this ability, we introduce a new benchmark, AmbigDocs. By leveraging Wikipedia's disambiguation pages, we identify a set of documents, belonging to different entities who share an ambiguous name. From these documents, we generate questions containing an ambiguous name and their corresponding sets of answers. Our analysis reveals that current state-of-the-art models often yield ambiguous answers or incorrectly merge information belonging to different entities. We establish an ontology categorizing four types of incomplete answers and automatic evaluation metrics to identify such categories. We lay the foundation for future work on reasoning across multiple documents with ambiguous entities.

Tesla Director of Artificial Intelligence and Autopilot Andrej Karpathy is leaving the company at a critical time - as it faces renewed probes over crashes and growing scrutiny. Tesla's head of artificial intelligence and autopilot Andrej Karpathy, pictured above at a conference, is leaving the company at a critical time'It's been a great pleasure to help Tesla towards its goals over the last 5 years and a difficult decision to part ways. In that time, Autopilot graduated from lane keeping to city streets and I look forward to seeing the exceptionally strong Autopilot team continue that momentum,' he wrote on Twitter, noting that he has no plans for what's next. Tesla CEO Elon Musk replied to thank him for his work at the company. The leadership change comes at a challenging time, as Tesla faces renewed scrutiny from US regulators over crashes involving drivers who used Autopilot and works to expand the latest version of Full Self Driving (FSD) to a larger number of customers.

Tables provide valuable knowledge that can be used to verify textual statements. While a number of works have considered table-based fact verification, direct alignments of tabular data with tokens in textual statements are rarely available. Moreover, training a generalized fact verification model requires abundant labeled training data. In this paper, we propose a novel system to address these problems. Inspired by counterfactual causality, our system identifies token-level salience in the statement with probing-based salience estimation. Salience estimation allows enhanced learning of fact verification from two perspectives. From one perspective, our system conducts masked salient token prediction to enhance the model for alignment and reasoning between the table and the statement. From the other perspective, our system applies salience-aware data augmentation to generate a more diverse set of training instances by replacing non-salient terms. Experimental results on TabFact show the effective improvement by the proposed salience-aware learning techniques, leading to the new SOTA performance on the benchmark. Our code is publicly available at https://github.com/luka-group/Salience-aware-Learning .

Elon Musk has been banging the drum for Tesla's with'Full Self-Driving' (FSD) for more than five years, but a number of leaked emails reveal the technology is far off from providing hands-free capabilities. Documents between Tesla attorneys and the California Department of Motor Vehicles (DMV) say vehicles using the firm's latest beta version, known as'Autosteer on City Streets' will not surpass Level 2 autonomy. This level of autonomy requires drivers to remain aware and control the brake, accelerator and steering - despite Musk promising'full self driving' by 2021. Attorneys for the carmaker said the FSD beta upgrade'does not make it autonomous under the DMV's definition,' along with stating the Level 2 of will'remain largely unchanged' in a full customer rollout. Elon Musk has been banging the drum for Tesla's with'Full Self-Driving' (FSD) for more than five years, but a number of leaked emails reveal the technology is far from providing hands-free capabilities'City Streets continues to firmly root the vehicle in SAE Level 2 capability and does not make it autonomous under the DMV's definition, wrote Eric Williams, Tesla associate general counsel, in a statement attached to an email with the California DMV that has been published to PlainSite.

In the immortal words of Arthur the Aardvark, "Havin' fun isn't hard, when you've got a library card!" But how much fun can you really have with a library card? Turns out, more than I expected. Libraries across America are adding video games to their collections available for checkout. Gamers with an incessant appetite for new experiences or anyone looking to play video games for free should contact their local library to see if they have a collection.

Tesla CEO Elon Musk announced the firm is rolling out'Full Self-Driving' beta to a limited amount of customers deemed'expert and careful drivers.' The upgrade is set to be released Tuesday, October 20 and provides access to specific features for highway assistance and automatic traffic light and stop sign control. The software is an advanced version of Autopilot, which means drivers must keep their hands on the wheel to use the technology. Telsa's rollout is setting the firm up to release one million robotaxi vehicles, which are set to be ready by this year. However, the fate of both Tesla's robotaxis and fully autonomous vehicles depend on regulatory approval.

Security cameras watching a highway in Taiwan captured the moment a white Tesla Model 3 vehicle plowing into truck that was rolled over on its side. Reports say the driver of the Tesla did not see the overturned Truck while cruising with the Autopilot driver assistant feature activated. The footage also shows that the car's emergency automatic braking system was applied at the last second, due to smoke coming from the tires moments before the collision. An image of the aftermath shows the entire front-end of the Tesla pierced through the roof of the truck, but reports note that neither of the drivers were injured. Tesla's Autopilot features allow the vehicle to steer, accelerate and brake automatically within a lane.

As people make efforts to stay apart from each other physically, video games are filling the socializing gap. As people make efforts to stay apart from each other physically, video games are filling the socializing gap. Some people look at the weeks ahead and wonder how they will keep themselves from going stir crazy. Across the U.S., new restrictions have limited in-person gatherings in an effort to stem the spread of coronavirus infection, as concern grows from watching its effects on the hard-hit populations of China and Italy, where thousands have died. But other Americans already have a plan to help combat social isolation: video games.

A research team at Ben-Gurion University have created a simple projection system able to trick Tesla's Autopilot into seeing things that aren't actually there. Using commercially available drones and a cheap projector - the kind a person might use to watch television in an apartment of small home - the team projected a series of deceptive images onto the road. The images included false traffic lines, a false speed limit sign, and an image of Elon Musk himself, projected on the road as if her were an endangered pedestrian. The researchers collectively labeled all these different visual phenomena as'phantoms,' according to a report in ArsTechnica. While the Tesla they tested reacted to every phantom in some way, most of its responses were fairly mild.

DETROIT - A Tesla Model S involved in a fatal crash with a semitrailer in Florida March 1 was operating on the company's semi-autonomous Autopilot system, federal investigators have determined. The car drove beneath the trailer, killing the driver, in a crash that is strikingly similar to one that happened on the other side of Florida in 2016 that also involved use of Autopilot. In both cases, neither the driver nor the Autopilot system stopped for the trailers, and the roofs of the cars were sheared off. The crash, which remains under investigation by the National Transportation Safety Board and the National Highway Traffic Safety Administration, raises questions about the effectiveness of Autopilot, which uses cameras, long-range radar and computers to detect objects in front of the cars to avoid collisions. The system also can keep a car in its lane, change lanes and navigate freeway interchanges.