PCWorld reports Microsoft is resuming forced M365 Copilot installations after a brief pause, targeting completion by July 1st for all Microsoft 365 users. Previous forced installations caused significant backlash and critical bugs, including Copilot accessing confidential emails, contributing to Windows 11's stagnating market share. IT administrators can opt out of automatic installation, while Microsoft tests an "uninstall AI bloat" button in Windows Insider builds despite continuing AI-driven OS plans. Between October 2025 and March 2026, many Microsoft 365 users discovered that the Copilot AI app was automatically installed on their computers. After heavy vocal backlash and several crucial bugs, like the one that allowed Copilot to read confidential emails, the forced installation of Microsoft 365 Copilot was suspended. However, Microsoft has now decided to reinstate the app's forced installation.

PCWorld highlights Apple's practical AI features launching this fall, including Visual Intelligence for scanning restaurant bills and splitting costs via Apple Cash. Call Context uses Apple Intelligence to automatically display account numbers and confirmation codes during customer service calls, eliminating awkward pauses. Apple Wallet now offers a "create a pass" tool that generates digital passes from physical cards using barcode and QR code scanning. Apple mostly steered clear of the big, dumb AI announcements we often see at the big tech keynotes -- you know, things like AI pop song generators, AI-generated avatars, and virtual AI try-on tools that make for great headlines but aren't all that practical. OK, so Apple did indulge in "look at this!" AI tomfoolery, like photo-retouch tools that can wipe groups of people out of the frame and a revamped Image Playground app that can essentially deep-fake your friends (ugh). That said, Apple also unleashed a series of smaller Apple Intelligence and Siri AI features (all due to arrive this fall) that might actually be useful on a daily basis. Here are five of my favorites.

Follow this section to personalize your feed and get instant alerts. Follow Go to your personalized feed WHY FOLLOW? Smart Alerts: Get notified about major news as it happens. Follow this tag to personalize your feed and get instant alerts. Follow Go to your personalized feed WHY FOLLOW?

We got a short trailer with no real gameplay at the latest Nintendo Direct. Nintendo just officially announced a remake of, ending months of rumors and speculation . The company dropped a short trailer at today's Nintendo Direct livestream and this looks like a top-to-bottom remake with a massive graphical facelift. The game launches later this year for Switch 2, with more details to come. There's still a lot we don't know, including price and if there will be any new content.

Descendants of these rodents are still alive today and are'like tiny Arctic pack rats.' More information Adding us as a Preferred Source in Google by using this link indicates that you would like to see more of our content in Google News results. Researchers document a cluster of ancient Arctic ground squirrel faecal pellets preserved in permafrost at Hunker Creek, Yukon, in August 2022. These coprolites contain remarkably intact ancient DNA, offering rare glimpses into ice age ecosystems. Breakthroughs, discoveries, and DIY tips sent six days a week.

PCWorld reports that Google's Manifest V3 update will permanently disable popular ad blockers like uBlock Origin in Chrome by late June. This transition from Manifest V2 aims to enhance Chrome's security and speed, but inadvertently limits ad blocker functionality as a side effect. Chrome 150 or 151 will likely remove all workarounds, forcing users to seek alternative browsers or accept reduced ad-blocking capabilities. Google has been working for some time on a way to block old browser extensions in Google Chrome. This goes hand in hand with the switch from Manifest V2 to Manifest V3, a newer and presumably more secure architecture for the popular browser. As early as March 2025, this rendered some extensions--including popular ad blockers such as uBlock Origin--suddenly unusable, even though it was still possible to access them with a workaround.

Visible-infrared person re-identification (VI-ReID) aims to match visible and infrared images of the same individual. Supervised VI-ReID (SVI-ReID) methods have achieved promising performance under the guidance of manually annotated identity labels. However, the substantial annotation cost severely limits their scalability in real-world applications. As a result, unsupervised VI-ReID (UVI-ReID) methods have attracted increasing attention. These methods typically rely on pseudo-labels generated by clustering and matching algorithms to replace manual annotations. Nevertheless, the quality of pseudo-labels is often difficult to guarantee, and low-quality pseudo-labels can significantly hinder model performance improvements.

The rapid advancement of large language models (LLMs) has blurred the line between AI-generated and human-written text. This progress brings societal risks such as misinformation, authorship ambiguity, and intellectual property concerns, highlighting the urgent need for reliable AI-generated text detection methods. However, recent advances in generative language modeling have resulted in significant overlap between the feature distributions of human-written and AI-generated text, blurring classification boundaries and making accurate detection increasingly challenging. To address the above challenges, we propose a DNA-inspired perspective, leveraging a repair-based process to directly and interpretably capture the intrinsic differences between human-written and AI-generated text.

Maintaining consistent model performance across domains is a fundamental challenge in machine learning. While recent work has explored using LLM-generated data for fine-tuning, its impact on cross-domain generalization remains poorly understood. This paper presents a systematic analysis revealing that fine-tuning with LLM-generated data not only improves target task performance but also reduces non-target task degradation compared to fine-tuning with ground truth data. Through analyzing the data sequence in tasks of various domains, we demonstrate that this enhancement of non-target task robustness stems from the reduction of high perplexity tokens found in LLM-generated sequences. Following our findings, we showed that masking high perplexity tokens in ground truth training data achieves similar non-target task performance preservation, comparable to using LLM-generated data. Extensive experiments across different model families and scales, including Gemma 2 IT 2B, Llama 3 8B Instruct, and three additional models, agree with our findings. To the best of our knowledge, this is the first work to provide an empirical explanation based on token perplexity reduction to mitigate catastrophic forgetting in LLMs after fine-tuning, offering valuable insights for developing more robust fine-tuning strategies.

Accurately modeling the target-drug complex at atom level presents a significant challenge in the computer-aided drug design. Traditional methods that rely solely on rigid transformations often fail to capture the adaptive interactions between targets and drugs, particularly during substantial conformational changes in targets upon ligand binding, which becomes especially critical when learning target-drug interactions in drug design. Accurately modeling these changes is crucial for understanding target-drug interactions and improving drug efficacy. To address these challenges, we introduce DynaPhArM, an SE(3)-Equivariant Transformer model specifically designed to capture adaptive alterations occurring within target-drug interactions. DynaPhArM utilizes the cooperative scalar-vector representation, drug-specific embeddings, and a diffusion process to effectively model the evolving dynamics of interactions between targets and drugs. Furthermore, we integrate physical information and energetic principles that maintain essential geometric constraints, such as bond lengths, bond angles, van der Waals forces (vdW), within a multi-task learning (MTL) framework to enhance accuracy. Experimental results demonstrate that DynaPhArM achieves state-of-the-art performance with an overall root mean square deviation (RMSD) of 2.01 Å and a sc-RMSD of 0.29 Å while exhibiting higher success rates compared to existing methodologies. Additionally, DynaPhArM shows promise in enhancing drug specificity, thereby simulating how targets adapt to various drugs through precise modeling of atomic-level interactions and conformational flexibility.