Despite the great potential of large language models(LLMs) in machine comprehension, it is still disturbing to fully count on them in real-world scenarios. This is probably because there is no rational explanation for whether the comprehension process of LLMs is aligned with that of experts. In this paper, we propose SCOP to carefully examine how LLMs perform during the comprehension process from a cognitive view. Specifically, it is equipped with a systematical definition of five requisite skills during the comprehension process, a strict framework to construct testing data for these skills, and a detailed analysis of advanced open-sourced and closed-sourced LLMs using the testing data. With SCOP, we find that it is still challenging for LLMs to perform an expert-level comprehension process. Even so, we notice that LLMs share some similarities with experts, e.g., performing better at comprehending local information than global information. Further analysis reveals that LLMs can be somewhat unreliable -- they might reach correct answers through flawed comprehension processes. Based on SCOP, we suggest that one direction for improving LLMs is to focus more on the comprehension process, ensuring all comprehension skills are thoroughly developed during training.

Federated Learning with client-level differential privacy (DP) provides a promising framework for collaboratively training models while rigorously protecting clients' privacy. However, classic approaches like DP-FedAvg struggle when clients have heterogeneous privacy requirements, as they must uniformly enforce the strictest privacy level across clients, leading to excessive DP noise and significant model utility degradation. Existing methods to improve the model utility in such heterogeneous privacy settings often assume a trusted server and are largely heuristic, resulting in suboptimal performance and lacking strong theoretical underpinnings. In this work, we address these challenges under a practical attack model where both clients and the server are honest-but-curious. We propose GDPFed, which partitions clients into groups based on their privacy budgets and achieves client-level DP within each group to reduce the privacy budget waste and hence improve the model utility. Based on the privacy and convergence analysis of GDPFed, we find that the magnitude of DP noise depends on both model dimensionality and the per-group client sampling ratios. To further improve the performance of GDPFed, we introduce GDPFed$^+$, which integrates model sparsification to eliminate unnecessary noise and optimizes per-group client sampling ratios to minimize convergence error. Extensive empirical evaluations on multiple benchmark datasets demonstrate the effectiveness of GDPFed$^+$, showing substantial performance gains compared with state-of-the-art methods.

Quantum Federated Learning (QFL) merges privacy-preserving federation with quantum computing gains, yet its resilience to adversarial noise is unknown. We first show that QFL is as fragile as centralized quantum learning. We propose Robust Quantum Federated Learning (RobQFL), embedding adversarial training directly into the federated loop. RobQFL exposes tunable axes: client coverage $γ$ (0-100\%), perturbation scheduling (fixed-$\varepsilon$ vs $\varepsilon$-mixes), and optimization (fine-tune vs scratch), and distils the resulting $γ\times \varepsilon$ surface into two metrics: Accuracy-Robustness Area and Robustness Volume. On 15-client simulations with MNIST and Fashion-MNIST, IID and Non-IID conditions, training only 20-50\% clients adversarially boosts $\varepsilon \leq 0.1$ accuracy $\sim$15 pp at $< 2$ pp clean-accuracy cost; fine-tuning adds 3-5 pp. With $\geq$75\% coverage, a moderate $\varepsilon$-mix is optimal, while high-$\varepsilon$ schedules help only at 100\% coverage. Label-sorted non-IID splits halve robustness, underscoring data heterogeneity as a dominant risk.

Large language models increasingly rely on synthetic data due to human-written content scarcity, yet recursive training on model-generated outputs leads to model collapse, a degenerative process threatening factual reliability. We define knowledge collapse as a distinct three-stage phenomenon where factual accuracy deteriorates while surface fluency persists, creating "confidently wrong" outputs that pose critical risks in accuracy-dependent domains. Through controlled experiments with recursive synthetic training, we demonstrate that collapse trajectory and timing depend critically on instruction format, distinguishing instruction-following collapse from traditional model collapse through its conditional, prompt-dependent nature. We propose domain-specific synthetic training as a targeted mitigation strategy that achieves substantial improvements in collapse resistance while maintaining computational efficiency. Our evaluation framework combines model-centric indicators with task-centric metrics to detect distinct degradation phases, enabling reproducible assessment of epistemic deterioration across different language models. These findings provide both theoretical insights into collapse dynamics and practical guidance for sustainable AI training in knowledge-intensive applications where accuracy is paramount.

Fox News chief political analyst Brit Hume explains why President Donald Trump should not remove himself from the peace negotiations between Russia and Ukraine and more on'Special Report.' When Russia invaded Ukraine in February 2022, many experts predicted Kyiv's quick fall. When Ukraine pushed back overextended Russian forces, the same experts confidently said that Russia's mass -- a population almost four times larger than Ukraine -- would certainly grind Ukraine down. Triumph for Putin was inevitable. But, an odd thing happened on the way to Russia's victory parade: Ukraine is outfighting Russia.

When I was growing up, I watched a lot of sci-fi movies about aliens that come to Earth. The extraterrestrials in popular culture, however, always looked so familiar that I found them far-fetched. What are the chances that E.T., the Predator, or ALF would develop arms and legs, a humanlike face, and opposable thumbs? Perhaps as a result, I associated alien life more with fantasy than with science, and I never gave much thought to what a visit would really look like. But my attitude started to change in 2020, when I read Liu Cixin's "The Three-Body Problem" and its two sequels.

At least three people have been killed, 18 wounded, and dozens of buildings set on fire in Kyiv, including the seat of the government, following a Russian drone and missile attack in Ukraine's capital, according to officials and local news reports. Kyiv Mayor Vitali Klitschko was initially quoted by Reuters news agency as saying that the attack early on Sunday killed an infant and a young woman, and sparked fires at several high-rise apartment buildings in the city's west and east. Medics were called to the leafy Darnytskyi district to the east of the Dnipro River, where a four-storey apartment building caught fire from the debris of drones destroyed in the overnight attack, Klitschko added. The Ukrainian news website Kyiv Independent also reported that an elderly woman also died in a shelter in the city's Darnytskyi district following the attack, although the cause of of her death was not immediately clear. The State Emergency Service confirmed at least one fatality in Kyiv and at least 18 injured.

Explosions were heard in cities across Ukraine, including Kyiv, Kharkiv and Dnipro, late on Saturday, as Russian forces launched another large-scale drone attack on the country, the Kyiv Independent reported, citing officials. A Russian strike in the town of Putyvl in Ukraine's northeastern Sumy region killed one person and wounded several others, regional Governor Oleh Hryhorov said. A nine-year-old child was among those injured. A separate Russian drone attack in Zaporizhia in the southeast also wounded at least 15 people, four of whom were hospitalised, said Ivan Fedorov, the head of the military administration in the region, which is partially occupied by Russia. Authorities in Ukraine's Chernihiv said a Russian drone dropped leaflets in the form of 100 Hryvnia bills, offering residents real money in exchange for coordinates to help Russia target Ukrainian forces.

Melania Trump urges parents to prepare their children for the growth of A.I. and argues the technology should be treated as if it were a child itself. First lady Melania Trump attends a meeting of the White House Task Force on Artificial Intelligence (AI) Education in the East Room at the White House in Washington, D.C., Sept. 4, 2025. FRONT AND CENTER: First lady Melania Trump hosted an artificial intelligence meeting with top industry leaders, including Google CEO Sundar Pichai Thursday, as she stressed the importance of managing AI's growth "responsibly." WORLD-CHANGING: If you were investing in the late 1990s, you'll remember the euphoria of the dot-com boom. Anything with a ".com" at the end of its name could raise millions in capital and see its stock price double or triple overnight.

When the chief executive of the Financial Times suggested at a media conference this summer that rival publishers might consider a "Nato for news" alliance to strengthen negotiations with artificial intelligence companies there was a ripple of chuckles from attendees. Yet Jon Slade's revelation that his website had seen a "pretty sudden and sustained" decline of 25% to 30% in traffic to its articles from readers arriving via internet search engines quickly made clear the serious nature of the threat the AI revolution poses. Queries typed into sites such as Google, which accounts for more than 90% of the search market, have been central to online journalism since its inception, with news providers optimising headlines and content to ensure a top ranking and revenue-raising clicks. But now Google's AI Overviews, which sit at the top of the results page and summarise responses and often negate the need to follow links to content, as well as its recently launched AI Mode tab that answers queries in a chatbot format, have prompted fears of a "Google zero" future where traffic referrals dry up. "This is the single biggest change to search I have seen in decades," says one senior editorial tech executive.