Measuring the generalization ability of Large Language Models (LLMs) is challenging due to data contamination. As models grow and computation becomes cheaper, ensuring tasks and test cases are unseen during training phases will become nearly impossible. We argue that knowledge-retrieval and reasoning tasks are not ideal for measuring generalization, as LLMs are not trained for specific tasks. Instead, we propose user behavior prediction, also a key aspect of personalization, as a theoretically sound, scalable, and robust alternative. We introduce a novel framework for this approach and test it on movie and music recommendation datasets for GPT-4o, GPT-4o-mini, and Llama-3.1-8B-Instruct. Results align with our framework's predictions, showing GPT-4o outperforms GPT-4o-mini and Llama, though all models have much room for improvement, especially Llama.

Recent advances in 3D Gaussian Splatting (3DGS) have enabled real-time novel view synthesis (NVS) with impressive quality in indoor scenes. However, achieving high-fidelity rendering requires meticulously captured images covering the entire scene, limiting accessibility for general users. We aim to develop a practical 3DGS-based NVS framework using simple panorama-style motion with a handheld camera (e.g., mobile device). While convenient, this rotation-dominant motion and narrow baseline make accurate camera pose and 3D point estimation challenging, especially in textureless indoor scenes. To address these challenges, we propose LighthouseGS, a novel framework inspired by the lighthouse-like sweeping motion of panoramic views. LighthouseGS leverages rough geometric priors, such as mobile device camera poses and monocular depth estimation, and utilizes the planar structures often found in indoor environments. We present a new initialization method called plane scaffold assembly to generate consistent 3D points on these structures, followed by a stable pruning strategy to enhance geometry and optimization stability. Additionally, we introduce geometric and photometric corrections to resolve inconsistencies from motion drift and auto-exposure in mobile devices. Tested on collected real and synthetic indoor scenes, LighthouseGS delivers photorealistic rendering, surpassing state-of-the-art methods and demonstrating the potential for panoramic view synthesis and object placement.

Nowadays, misinformation articles, especially multimodal ones, are widely spread on social media platforms and cause serious negative effects. To control their propagation, Multimodal Misinformation Detection (MMD) becomes an active topic in the community to automatically identify misinformation. Previous MMD methods focus on supervising detectors by collecting offline data. However, in real-world scenarios, new events always continually emerge, making MMD models trained on offline data consistently outdated and ineffective. To address this issue, training MMD models under online data streams is an alternative, inducing an emerging task named continual MMD. Unfortunately, it is hindered by two major challenges. First, training on new data consistently decreases the detection performance on past data, named past knowledge forgetting. Second, the social environment constantly evolves over time, affecting the generalization on future data. To alleviate these challenges, we propose to remember past knowledge by isolating interference between event-specific parameters with a Dirichlet process-based mixture-of-expert structure, and anticipate future environmental distributions by learning a continuous-time dynamics model. Accordingly, we induce a new continual MMD method DAEDCMD. Extensive experiments demonstrate that DAEDCMD can consistently and significantly outperform the compared methods, including six MMD baselines and three continual learning methods.

Humans can imagine various atmospheres and settings when listening to music, envisioning movie scenes that complement each piece. For example, slow, melancholic music might evoke scenes of heartbreak, while upbeat melodies suggest celebration. This paper explores whether a Music Language Model, e.g. MU-LLaMA, can perform a similar task, called Music Scene Imagination (MSI), which requires cross-modal information from video and music to train. To improve upon existing music captioning models which focusing solely on musical elements, we introduce MusiScene, a music captioning model designed to imagine scenes that complement each music. In this paper, (1) we construct a large-scale video-audio caption dataset with 3,371 pairs, (2) we finetune Music Understanding LLaMA for the MSI task to create MusiScene, and (3) we conduct comprehensive evaluations and prove that our MusiScene is more capable of generating contextually relevant captions compared to MU-LLaMA. We leverage the generated MSI captions to enhance Video Background Music Generation (VBMG) from text.

The exponential growth of social media and generative AI has transformed information dissemination, fostering connectivity but also accelerating the spread of misinformation. Understanding information propagation dynamics and developing effective control strategies is essential to mitigate harmful content. Traditional models, such as SIR, provide basic insights but inadequately capture the complexities of online interactions. Advanced methods, including attention mechanisms and graph neural networks, enhance accuracy but typically overlook user psychology and behavioral dynamics. Large language models (LLMs), with their human-like reasoning, offer new potential for simulating psychological aspects of information spread. We introduce an LLM-based simulation environment capturing agents' evolving attitudes, emotions, and responses. Initial experiments, however, revealed significant gaps between LLM-generated behaviors and authentic human dynamics, especially in stance detection and psychological realism. A detailed evaluation through Social Information Processing Theory identified major discrepancies in goal-setting and feedback evaluation, stemming from the lack of emotional processing in standard LLM training. To address these issues, we propose the Social Information Processing-based Chain of Thought (SIP-CoT) mechanism enhanced by emotion-guided memory. This method improves the interpretation of social cues, personalization of goals, and evaluation of feedback. Experimental results confirm that SIP-CoT-enhanced LLM agents more effectively process social information, demonstrating behaviors, attitudes, and emotions closer to real human interactions. In summary, this research highlights critical limitations in current LLM-based propagation simulations and demonstrates how integrating SIP-CoT and emotional memory significantly enhances the social intelligence and realism of LLM agents.

Amazon Prime Day is now almost a week. Prime Day started today and will go on for three more days. We'll be dangerously caffeinated and working shifts 20 hours a day from now through Friday, July 11. The WIRED Reviews team has been prepping for weeks to bring you real savings on the very best tech, and we only recommend products we've actually tested and approved. If you're looking for up-to-the-minute coverage of check out our Amazon Prime Day liveblog, which will run from 5 am to midnight daily. Deals on computers, routers, monitors, tablets, keyboards, and more. The Google Pixel Tablet (7/10, WIRED Recommends) is a good Android tablet. Where it really shines, though, is its ability to be paired with the charging speaker dock to transform into a smart speaker when you aren't using it as a tablet. Right now, only the tablet version is on sale, and it's a good price if you want to buy it for the sharp screen and overall solid performance. Enjoy simple, set-and-forget Wi-Fi courtesy of Amazon's Eero mesh systems. The tri-band Eero Pro 6E (7/10, WIRED Recommends) adds 6 GHz to the familiar 2.4-GHz and 5-GHz bands, for fast and dependable Wi-Fi. The Eero Plus subscription is expensive ( 10 per month or 100 per year) but includes comprehensive parental controls, advanced security, ad blocking, and even a password manager and VPN service. As the budget pick in our best mesh Wi-Fi systems guide, the Deco X20 is already a bargain. This Wi-Fi 6 dual-band mesh (2.4-GHz and 5-GHz) is easy to set up and delivered solid results in my tests. It's not the speediest mesh, but if your internet connection is 500 Mbps or less, it's likely enough. Each router has two gigabit Ethernet ports, and the vaselike design blends in easily on shelves or tables. This tri-band Wi-Fi 6E mesh system from TP-Link scores a place in our best mesh Wi-Fi systems guide. Easy to set up and configure through the mobile app, each unit has one 2.5 Gbps Ethernet port and two gigabit ports. It offers fast speeds at close range on the 6-GHz band, but was also fast on 5 GHz, and offered a decent range on 2.4 GHz. There are optional subscriptions for parental controls and enhanced security. Cheap laptops don't have to be terrible, and the HP Chromebook Plus x360 proves it. While there are more premium Chromebooks out there, none sell for so little on discount. It has a 14-inch 1080p screen, and unlike some Chromebooks at this price, it also comes with enough RAM and storage.

There are many filmmakers in Hollywood who view AI as antithetical to their creative process. This tension played a major role during the Hollywood strikes in 2023, with many on the picket lines expressing fears about job loss via automation. Talukdar, conversely, argues that AI tools will actually create new types of jobs, and enable studios to push their budgets further rather than slashing them. "There's this idea that instead of spending 50 million on a movie, you can now do it for 5 million, and there's some truth in that," he says. "But the other way to think about it--which is how every studio that we talked to is thinking about it--is now for that 50 million and for the same 100 people on that project, they're just going to be able to do what would have cost them 100 million before," he says.

In March, NewsGuard – a company that tracks misinformation – published a report claiming that generative Artificial Intelligence (AI) tools, such as ChatGPT, were amplifying Russian disinformation. NewsGuard tested leading chatbots using prompts based on stories from the Pravda network – a group of pro-Kremlin websites mimicking legitimate outlets, first identified by the French agency Viginum. The results were alarming: Chatbots "repeated false narratives laundered by the Pravda network 33 percent of the time", the report said. The Pravda network, which has a rather small audience, has long puzzled researchers. Some believe that its aim was performative – to signal Russia's influence to Western observers.

Amazon Prime Day is here, and the deals are dropping like the bass at a Skrillex show. This year, Amazon has expanded the event to four days, with most of the best Prime Day deals launching in the early hours of Tuesday morning and sitting on the metaphorical shelf through 11:59 pm PT on Friday, July 11. There are tens of thousands of products on sale this week, but comparatively few deserve your time and money. The WIRED Reviews team spends weeks prepping for Prime Day and will be working in shifts for 20 hours a day throughout the event to keep our coverage updated--all so you can score real savings on products we've tested and approved. If you're looking for up-to-the-minute coverage of lightning deals, this year's trending products, and fast sellers, along with what will surely be increasingly unhinged commentary, check out our Amazon Prime Day liveblog, which will run from 5 am to midnight daily. How Does WIRED Spot Deals? We start searching for the best Prime Day deals weeks ...

The introduction of ChatGPT has garnered significant attention within the NLP community and beyond. Previous studies have demonstrated ChatGPT's substantial advancements across various downstream NLP tasks, highlighting its adaptability and potential to revolutionize language-related applications. However, its capabilities and limitations in genre prediction remain unclear. This work analyzes three Large Language Models (LLMs) using the MovieLens-100K dataset to assess their genre prediction capabilities. Our findings show that ChatGPT, without fine-tuning, outperformed other LLMs, and fine-tuned ChatGPT performed best overall. We set up zero-shot and few-shot prompts using audio transcripts/subtitles from movie trailers in the MovieLens-100K dataset, covering 1682 movies of 18 genres, where each movie can have multiple genres. Additionally, we extended our study by extracting IMDb movie posters to utilize a Vision Language Model (VLM) with prompts for poster information. This fine-grained information was used to enhance existing LLM prompts. In conclusion, our study reveals ChatGPT's remarkable genre prediction capabilities, surpassing other language models. The integration of VLM further enhances our findings, showcasing ChatGPT's potential for content-related applications by incorporating visual information from movie posters.