To fix ideas, consider ECG recordings from healthy patients during rest.

Recent studies in long video understanding have harnessed the advanced visual-language reasoning capabilities of Large Multimodal Models (LMMs), driving the evolution of video-LMMs specialized for processing extended video sequences. However, the scalability of these models is severely limited by the overwhelming volume of visual tokens generated from extended video sequences. To address this challenge, this paper proposes FLoC, an efficient visual token compression framework based on the facility location function, a principled approach that swiftly selects a compact yet highly representative and diverse subset of visual tokens within a predefined budget on the number of visual tokens. By integrating the lazy greedy algorithm, our method achieves remarkable efficiency gains by swiftly selecting a compact subset of tokens, drastically reducing the number of visual tokens while guaranteeing near-optimal performance. Notably, our approach is training-free, model-agnostic, and query-agnostic, providing a versatile solution that seamlessly integrates with diverse video-LLMs and existing workflows. Extensive evaluations on large-scale benchmarks, such as Video-MME, MLVU, and LongVideoBench, demonstrate that our framework consistently surpasses recent compression techniques, highlighting not only its effectiveness and robustness in addressing the critical challenges of long video understanding, but also its efficiency in processing speed.

The changing dynamics of the digital world have led to several privacy challenges for businesses, large and small. This is placing increasing pressure on them to evolve their processes and strategies. Much of the burden stems from the sheer volume of data present today, and in fact, the volume of data is predicted to balloon to 175 zettabytes (ZB) by 2025. Today, it is simply beyond human capability to effectively process and protect privacy without the assistance of privacy-enhancing technologies (PETs). This has led to an explosion of adaptive machine learning (ML) algorithms that can wade through the mountain of data while continuously and efficiently changing their behavior in real-time as new data streams are fed into them.

As third party cookies are leaving our life, Google offers marketers Federated Learning of Cohorts (FLoC) so they can keep optimizing their ad spend. FLoC is a new tracking technology that is planned to be rolled out as third party cookies are banned by Google in 2023. FLoC uses federated learning principles. FLoC creates groups or "cohorts" by using browser historical data of users. Google has just started to test FLoC in 2021 in some countries.

Privacy-centric search engine DuckDuckGo has denounced Google's plans to implement an alternative to third-party browser cookies, which for years have allowed advertisers to track the activity of web users. The company has revealed it will block Federated Learning of Cohorts (FLoC), the proposed replacement, via both its search engine and Chrome browser extension. The DuckDuckGo search engine will disable FLoC functionality with immediate effect, while an update to the browser extension will give users the chance to opt out manually. However, the extension update is currently under review by Google, which vets all new additions to the Chrome Web Store. According to Google, the introduction of FLoC will allow people to benefit from greater levels of online privacy, but without destroying the underlying economics of the web.

Google is Google because of its lucrative advertising business--and that business works by letting advertisers target users based on what they do on the web. On Wednesday, Google announced what some observers have framed as a major shift in that setup: The company's Chrome browser will soon stop tracking individual users across different websites in order to serve them ads. While the change does allow the web giant and its advertising customers to continue tracking users to a certain extent, this appears to be a significant step away from Google's traditional model. David Temkin, Google's director of product management for ads privacy and trust, described the decision as a move to address growing concerns about digital privacy. "People shouldn't have to accept being tracked across the web in order to get the benefits of relevant advertising," he wrote in a blog post announcing the change. "And advertisers don't need to track individual consumers across the web to get the performance benefits of digital advertising."