Real-time personalization has advanced significantly in recent years, with platforms utilizing machine learning models to predict user preferences based on rich behavioral data on each individual user. Traditional approaches usually rely on embedding-based machine learning models to capture user preferences, and then reduce the final optimization task to nearest-neighbors, which can be performed extremely fast. However, these models struggle to capture complex user behaviors, which are essential for making accurate recommendations. Transformer-based models, on the other hand, are known for their practical ability to model sequential behaviors, and hence have been intensively used in personalization recently to overcome these limitations. However, optimizing recommendations under transformer-based models is challenging due to their complicated architectures. In this paper, we address this challenge by considering a specific class of transformers, showing its ability to represent complex user preferences, and developing efficient algorithms for real-time personalization. We focus on a particular set of transformers, called simple transformers, which contain a single self-attention layer. We show that simple transformers are capable of capturing complex user preferences. We then develop an algorithm that enables fast optimization of recommendation tasks based on simple transformers. Our algorithm achieves near-optimal performance in sub-linear time. Finally, we demonstrate the effectiveness of our approach through an empirical study on datasets from Spotify and Trivago. Our experiment results show that (1) simple transformers can model/predict user preferences substantially more accurately than non-transformer models and nearly as accurately as more complex transformers, and (2) our algorithm completes simple-transformer-based recommendation tasks quickly and effectively.

Social media users and inauthentic accounts, such as bots, may coordinate in promoting their topics. Such topics may give the impression that they are organically popular among the public, even though they are astroturfing campaigns that are centrally managed. It is challenging to predict if a topic is organic or a coordinated campaign due to the lack of reliable ground truth. In this paper, we create such ground truth by detecting the campaigns promoted by ephemeral astroturfing attacks. These attacks push any topic to Twitter's (X) trends list by employing bots that tweet in a coordinated manner in a short period and then immediately delete their tweets. We manually curate a dataset of organic Twitter trends. We then create engagement networks out of these datasets which can serve as a challenging testbed for graph classification task to distinguish between campaigns and organic trends. Engagement networks consist of users as nodes and engagements as edges (retweets, replies, and quotes) between users. We release the engagement networks for 179 campaigns and 135 non-campaigns, and also provide finer-grain labels to characterize the type of the campaigns and non-campaigns. Our dataset, LEN (Large Engagement Networks), is available in the URL below. In comparison to traditional graph classification datasets, which are small with tens of nodes and hundreds of edges at most, graphs in LEN are larger. The average graph in LEN has ~11K nodes and ~23K edges. We show that state-of-the-art GNN methods give only mediocre results for campaign vs. non-campaign and campaign type classification on LEN. LEN offers a unique and challenging playfield for the graph classification problem. We believe that LEN will help advance the frontiers of graph classification techniques on large networks and also provide an interesting use case in terms of distinguishing coordinated campaigns and organic trends.

Existing Existing automatic audio generation methods struggle to generate podcast-like audio programs effectively. The key challenges lie in in-depth content generation, appropriate and expressive voice production. This paper proposed PodAgent, a comprehensive framework for creating audio programs. PodAgent 1) generates informative topic-discussion content by designing a Host-Guest-Writer multi-agent collaboration system, 2) builds a voice pool for suitable voice-role matching and 3) utilizes LLM-enhanced speech synthesis method to generate expressive conversational speech. Given the absence of standardized evaluation criteria for podcast-like audio generation, we developed comprehensive assessment guidelines to effectively evaluate the model's performance. Experimental results demonstrate PodAgent's effectiveness, significantly surpassing direct GPT-4 generation in topic-discussion dialogue content, achieving an 87.4% voice-matching accuracy, and producing more expressive speech through LLM-guided synthesis. Demo page: https://podcast-agent.github.io/demo/. Source code: https://github.com/yujxx/PodAgent.

We have seen remarkable success in representation learning and language models (LMs) using deep neural networks. Many studies aim to build the underlying connections among different modalities via the alignment and mappings at the token or embedding level, but so far, most methods are very data-hungry, limiting their performance in domains such as music where paired data are less abundant. We argue that the embedding alignment is only at the surface level of multimodal alignment. In this paper, we propose a grand challenge of \textit{language model mapping} (LMM), i.e., how to map the essence implied in the LM of one domain to the LM of another domain under the assumption that LMs of different modalities are tracking the same underlying phenomena. We first introduce a basic setup of LMM, highlighting the goal to unveil a deeper aspect of cross-modal alignment as well as to achieve more sample-efficiency learning. We then discuss why music is an ideal domain in which to conduct LMM research. After that, we connect LMM in music with a more general and challenging scientific problem of \textit{learning to take actions based on both sensory input and abstract symbols}, and in the end, present an advanced version of the challenge problem setup.

Polarization is defined as divisive opinions held by two or more groups on substantive issues. As the world's third-largest democracy, Indonesia faces growing concerns about the interplay between political polarization and online toxicity, which is often directed at vulnerable minority groups. Despite the importance of this issue, previous NLP research has not fully explored the relationship between toxicity and polarization. To bridge this gap, we present a novel multi-label Indonesian dataset that incorporates toxicity, polarization, and annotator demographic information. Benchmarking this dataset using BERT-base models and large language models (LLMs) shows that polarization information enhances toxicity classification, and vice versa. Furthermore, providing demographic information significantly improves the performance of polarization classification.

We propose BGM2Pose, a non-invasive 3D human pose estimation method using arbitrary music (e.g., background music) as active sensing signals. Unlike existing approaches that significantly limit practicality by employing intrusive chirp signals within the audible range, our method utilizes natural music that causes minimal discomfort to humans. Estimating human poses from standard music presents significant challenges. In contrast to sound sources specifically designed for measurement, regular music varies in both volume and pitch. These dynamic changes in signals caused by music are inevitably mixed with alterations in the sound field resulting from human motion, making it hard to extract reliable cues for pose estimation. To address these challenges, BGM2Pose introduces a Contrastive Pose Extraction Module that employs contrastive learning and hard negative sampling to eliminate musical components from the recorded data, isolating the pose information. Additionally, we propose a Frequency-wise Attention Module that enables the model to focus on subtle acoustic variations attributable to human movement by dynamically computing attention across frequency bands. Experiments suggest that our method outperforms the existing methods, demonstrating substantial potential for real-world applications. Our datasets and code will be made publicly available.

From disinformation spread by AI chatbots to AI recommendations that inadvertently reinforce stereotypes, textual bias poses a significant challenge to the trustworthiness of large language models (LLMs). In this paper, we propose a multi-agent framework that systematically identifies biases by disentangling each statement as fact or opinion, assigning a bias intensity score, and providing concise, factual justifications. Evaluated on 1,500 samples from the WikiNPOV dataset, the framework achieves 84.9% accuracy$\unicode{x2014}$an improvement of 13.0% over the zero-shot baseline$\unicode{x2014}$demonstrating the efficacy of explicitly modeling fact versus opinion prior to quantifying bias intensity. By combining enhanced detection accuracy with interpretable explanations, this approach sets a foundation for promoting fairness and accountability in modern language models.

The best webcam makes you look your best -- for less. Most laptops ship with a 720p webcam, so a 1080p webcam or even a 4K webcam can make you stand out by comparison. I haven't tested every one of these webcams, just most of them -- and I describe how I test under my recommendations. You can also refer to my separate story on the best Windows Hello webcams for recommendations that aid videoconferencing and log you into your PC. Two recommendations from that list -- a premium and budget option -- appear on this list. Why you should trust me: PCWorld has been testing PC hardware since the 1980s, and I've been a technology journalist for 30 years, with extensive experience in reviewing PC hardware. When it comes to webcams, my recommendations are based on hands-on testing in real-world scenarios. I also tap the expertise and published reviews of my counterparts in Sweden and Germany to offer a comprehensive view of the current webcam landscape. The Anker PowerConf C200 offers an affordable yet substantial upgrade from the 720p cameras found in most laptop webcams to a 1440p webcam, yet does so at an affordable price.

The keyword for the iPhone 16e seems to be "compromise." In this episode, Devindra chats with Cherlynn about her iPhone 16e review and try to figure out who this phone is actually for. Also, they dive into Amazon's Alexa event, where we finally learned more about the company's AI-powered voice assistant. Alexa seems useful, but can we trust it? Listen below or subscribe on your podcast app of choice. If you've got suggestions or topics you'd like covered on the show, be sure to email us or drop a note in the comments! And be sure to check out our other podcast, Engadget News! Framework unveils a cheap 2-in-1 laptop and a…modular desktop? Devindra: This week, it's the iPhone 16e, which Cherlynn has reviewed. We're going to get her full thoughts on that thing. And also, Amazon held an AI event this week. We expected a lot of devices, but they spent 75 minutes talking about Alexa plus, which is the AI powered Alexa. Cherlynn: we expected a lot of devices. Cherlynn: one, at least one it's been a while. Devindra: Mr. Panos Panay was there, the father of the service and no devices, just him talking about AI. Cherlynn: Oh, and stay tuned at the end of this episode. Uh, I, we included an interview that I did with, um, the vice president of Alexa to talk more about the new Alexa plus. Devindra: Anyway, folks, if you're enjoying the show, please be sure to subscribe to us on iTunes or your podcaster of choice, leave us a review on iTunes and drop us an email at podcast@engadget.com. You can also join us on our live [00:01:00] stream on Thursday mornings, typically around 11 a. m. Um, you'll see our faces. Sometimes we'll do Q& A and show off devices as well. This week, uh, Sherilyn has the iPhone 16e, which is the least, um, impressive thing to show off. It's just like, Hey, you have an iPhone from 10 years ago, five, a while ago, Devindra: last, was there a single camera back iPhone? Cherlynn: Oh God, before that was 11. So, you know, it's like a flashback. So let's talk about this thing, Sherlynn. And I checked out your review. First of all, you gave it a really, um, I think serviceable score. Your title is what's your acceptable compromise. And really when we were talking about it last week, it really was like compromise seemed like the key word. The thing we kept coming back to was like just one camera, no mag safe, no fast wireless charging. What are your overall thoughts on this thing? Cherlynn: I mean, so that headline is like all thanks to our EIC, Aaron [00:02:00]Souppouris, because I was like, where, where do I go from here? How do I, so, so he's right. It is like, instead of what's in your wallet, it's like, what are you willing to take out your wallet? I'll tell you the story. So yesterday I was at the Amazon devices and services event where there were no devices and A bunch of other reporters had gathered and we were all like, you know, the, like, review's going up soon, right?

If you want to know why Tyler Perry put an 800m ( 635m) expansion of his studio complex on hold, type "two people in a living room in the mountains" into OpenAI's video generation tool. The result from artificial intelligence-powered Sora, which was released in the UK and Europe on Friday, indicates why the US TV and film mogul paused his plans. Perry said last year after seeing previews of Sora that if he wanted to produce that mountain shot, he may not need to build sets on location or on his lot. "I can sit in an office and do this with a computer, which is shocking to me," he said. The result from a simple text prompt is only five seconds long – you can go to up to 20 seconds and also stitch together much longer videos from the tool – and the "actors" display telltale problems with their hands (a common problem with AI tools).