--Foreign information operations conducted by Russian and Chinese actors exploit the United States' permissive information environment. These campaigns threaten democratic institutions and the broader Westphalian model. Y et, existing detection and mitigation strategies often fail to identify active information campaigns in real time. This paper introduces ChestyBot, a pragmatics-based language model that detects unlabeled foreign malign influence tweets with up to 98.34% accuracy. The model supports a novel framework to disrupt foreign influence operations in their formative stages. Foreign influence campaigns--particularly those attributed to Russia during the 2016 U.S. Presidential Election--demonstrated how state-sponsored social media operations can destabilize democratic societies [1]. During that campaign, social media posts emanating from one state - Russia - probably represented an intentional effort to influence the internal affairs of another country - the United States. Though these efforts may not have changed election outcomes, they nonetheless constitute an erosion of the Westphalian state model itself [2]. In recent years, China has attempted to use social media to influence foreign perceptions of internal matters such as the Beijing 2022 Winter Olympics, the origins of COVID-19, and the human rights abuses in Xinjiang [3]. Despite these initiatives, China has (as far as we can tell at the time of this writing) not performed a successful large-scale disinformation campaign directed against U.S. internal interests.

The rapid spread of online disinformation presents a global challenge, and machine learning has been widely explored as a potential solution. However, multilingual settings and low-resource languages are often neglected in this field. To address this gap, we conducted a shared task on multilingual claim retrieval at SemEval 2025, aimed at identifying fact-checked claims that match newly encountered claims expressed in social media posts across different languages. The task includes two subtracks: (1) a monolingual track, where social posts and claims are in the same language, and (2) a crosslingual track, where social posts and claims might be in different languages. A total of 179 participants registered for the task contributing to 52 test submissions. 23 out of 31 teams have submitted their system papers. In this paper, we report the best-performing systems as well as the most common and the most effective approaches across both subtracks. This shared task, along with its dataset and participating systems, provides valuable insights into multilingual claim retrieval and automated fact-checking, supporting future research in this field.

Effective information searching is essential for enhancing the reasoning and generation capabilities of large language models (LLMs). Recent research has explored using reinforcement learning (RL) to improve LLMs' search capabilities by interacting with live search engines in real-world environments. While these approaches show promising results, they face two major challenges: (1) Uncontrolled Document Quality: The quality of documents returned by search engines is often unpredictable, introducing noise and instability into the training process. (2) Prohibitively High API Costs: RL training requires frequent rollouts, potentially involving hundreds of thousands of search requests, which incur substantial API expenses and severely constrain scalability. To address these challenges, we introduce ZeroSearch, a novel RL framework that incentivizes the capabilities of LLMs to use a real search engine with simulated searches during training. Our approach begins with lightweight supervised fine-tuning to transform the LLM into a retrieval module capable of generating both useful and noisy documents in response to a query. During RL training, we employ a curriculum-based rollout strategy that incrementally degrades the quality of generated documents, progressively eliciting the model's reasoning ability by exposing it to increasingly challenging retrieval scenarios. Extensive experiments demonstrate that ZeroSearch effectively incentivizes the search capabilities of LLMs using a 3B LLM as the retrieval module. Remarkably, a 7B retrieval module achieves comparable performance to the real search engine, while a 14B retrieval module even surpasses it. Furthermore, it generalizes well across both base and instruction-tuned models of various parameter sizes and is compatible with a wide range of RL algorithms.

In an interview on BBC One's Sunday with Laura Kuenssberg programme, John said the government was on course to "rob young people of their legacy and their income", adding: "It's a criminal offence, I think. The government are just being absolute losers, and I'm very angry about it." Last week, Kyle was accused of being too close to big tech after analysis showed a sharp increase in his department's meetings with companies such as Google, Amazon, Apple and Meta since Labour won the election last July. John referred to a similar amendment that received peers' support last week, only to be removed by the government in the Commons, in a tit-for-tat process that threatens to mire the data bill. "It's criminal, in that I feel incredibly betrayed: the House of Lords did a vote, and it was more than two to one in our favour, the government just looked at it as if to say: 'Hmmm, well the old people … like me can afford it," said John.

Although their specs and features are updated every year, Apple's iPhones maintain the same general size and shape. But according to a new report, the tech giant is preparing a radical new form factor for one of its upcoming handsets. Apple tipster Mark Gurman claims the trillion-dollar tech company is working on a'mostly glass, curved iPhone'. The device will come'without any cutouts in the display', he claims, such as a notch at the top or a small circle for a front-facing camera. It will hit the shelves in a couple of years to mark 20 years since the very first iPhone went on sale – June 29, 2007.

Having your favourite TV show or movie interrupted by adverts is already frustrating, but things could soon be getting worse for Netflix users. At its'Upfront' event on Wednesday, the streaming giant revealed that it would be incorporating adverts made with'generative AI'. Arriving in 2026, these AI-generated adverts will begin to appear not only during mid-content breaks but also when users press pause. And the only way to get rid of these annoying intrusions will be to pay for the more expensive ad-free subscriptions. But in a further twist, Netflix says AI would be used'instantly marry advertisers' ads with the worlds of our shows'.

Larry Niven is one of the biggest names in the history of science fiction, and it was a privilege to interview him via Zoom at his home in Los Angeles recently. His 1970 novel Ringworld is the latest pick for the New Scientist Book Club, but he has also written a whole space-fleet-load of novels and short stories over the years, including my favourite sci-fi of all time, A World Out of Time. At 87 years of age, he is very much still writing. I spoke to him about Ringworld, his start in sci-fi, his favourite work over the years, his current projects and whether he thinks humankind will ever leave this solar system. This is an edited version of our conversation.

Real-world RAG applications often encounter long-context input scenarios, where redundant information and noise results in higher inference costs and reduced performance. To address these challenges, we propose LongRefiner, an efficient plug-and-play refiner that leverages the inherent structural characteristics of long documents. LongRefiner employs dual-level query analysis, hierarchical document structuring, and adaptive refinement through multi-task learning on a single foundation model. Experiments on seven QA datasets demonstrate that LongRefiner achieves competitive performance in various scenarios while using 10x fewer computational costs and latency compared to the best baseline. Further analysis validates that LongRefiner is scalable, efficient, and effective, providing practical insights for real-world long-text RAG applications. Our code is available at https://github.com/ignorejjj/LongRefiner.

Modern edge devices, such as cameras, drones, and Internet-of-Things nodes, rely on deep learning to enable a wide range of intelligent applications, including object recognition, environment perception, and autonomous navigation. However, deploying deep learning models directly on the often resource-constrained edge devices demands significant memory footprints and computational power for real-time inference using traditional digital computing architectures. In this paper, we present WISE, a novel computing architecture for wireless edge networks designed to overcome energy constraints in deep learning inference. WISE achieves this goal through two key innovations: disaggregated model access via wireless broadcasting and in-physics computation of general complex-valued matrix-vector multiplications directly at radio frequency. Using a software-defined radio platform with wirelessly broadcast model weights over the air, we demonstrate that WISE achieves 95.7% image classification accuracy with ultra-low operation power of 6.0 fJ/MAC per client, corresponding to a computation efficiency of 165.8 TOPS/W. This approach enables energy-efficient deep learning inference on wirelessly connected edge devices, achieving more than two orders of magnitude improvement in efficiency compared to traditional digital computing.

Large Language Models (LLMs) have become increasingly central to recommendation scenarios due to their remarkable natural language understanding and generation capabilities. Although significant research has explored the use of LLMs for various recommendation tasks, little effort has been dedicated to verifying whether they have memorized public recommendation dataset as part of their training data. This is undesirable because memorization reduces the generalizability of research findings, as benchmarking on memorized datasets does not guarantee generalization to unseen datasets. Furthermore, memorization can amplify biases, for example, some popular items may be recommended more frequently than others. In this work, we investigate whether LLMs have memorized public recommendation datasets. Specifically, we examine two model families (GPT and Llama) across multiple sizes, focusing on one of the most widely used dataset in recommender systems: MovieLens-1M. First, we define dataset memorization as the extent to which item attributes, user profiles, and user-item interactions can be retrieved by prompting the LLMs. Second, we analyze the impact of memorization on recommendation performance. Lastly, we examine whether memorization varies across model families and model sizes. Our results reveal that all models exhibit some degree of memorization of MovieLens-1M, and that recommendation performance is related to the extent of memorization. We have made all the code publicly available at: https://github.com/sisinflab/LLM-MemoryInspector