Brief

SAP is emphasizing AI sovereignty, particularly for its European customers concerned about geopolitical instability and data control. The company is partnering with European AI providers like Mistral and n8n, in addition to previously featured partners like Anthropic, to offer AI solutions hosted within European data centers. This move aims to address customer demands for localized data storage and the use of regional AI companies, positioning SAP's Autonomous Enterprise vision as a response to these growing concerns. AI

Automakers are undergoing a significant shift in vehicle architecture, moving from numerous distributed Electronic Control Units (ECUs) to a centralized, single-brain system. This transformation, driven by the rise of software-defined vehicles, advanced driver-assistance systems, and electrification, aims to reduce complexity, weight, and cost. The new E/E architecture will enable vehicles to be updated over-the-air, turning them into long-term platforms for recurring revenue rather than one-time product sales. AI

Researchers have introduced PolyILR, a novel method for decomposing compositional data that accounts for hierarchical structures. This technique creates a canonical orthonormal decomposition of the Aitchison tangent space, aligning with any tree topology. PolyILR yields stable and interpretable features, enabling inference at various tree resolutions and showing potential applications in probabilistic modeling. AI

Hong Kong's first five-year plan is anticipated to be released by the end of the third quarter, ahead of the original year-end target. This accelerated timeline is attributed to strong collaboration across various sectors. The public consultation for this plan is scheduled to commence on June 15th. AI

China's Cyberspace Administration has guided key websites and platforms to establish a self-regulatory convention aimed at improving the online business environment. This convention mandates platforms to actively remove unverified false information concerning businesses and protect entrepreneurs' rights. It also includes measures to manage trending topics, optimize recommendation algorithms, and penalize accounts that repeatedly spread negative business-related content. AI

Xingneng Xuanguang, a company focused on controlled nuclear fusion FRC technology, has successfully completed two rounds of Series A financing, raising a total of 500 million yuan. This funding will be allocated to further enhance and advance the parameters of their high-performance FRC devices. Separately, Liipus LiiPuS secured tens of millions of yuan in seed and angel rounds for its high-end marine integrated electric propulsion systems. AI

Researchers have developed CRUMB, a novel inference wrapper designed to improve the efficiency of prior-fitted networks (PFNs). PFNs are powerful tabular foundation models that can perform in-context learning, but their self-attention mechanisms lead to computationally expensive inference with large datasets. CRUMB addresses this by clustering test queries, selecting distributionally matched training subsets using MMD minimization, and then performing inference on these reduced batches. This method is architecture-agnostic and has demonstrated superior performance on the TabArena benchmark compared to existing context selection strategies, while also showing resilience to covariate drift. AI

LiiPuS, a company specializing in high-end marine integrated electric propulsion systems, has secured tens of millions of yuan in seed and angel funding. The funds will support prototype testing, product refinement, and small-batch delivery, aiming to accelerate the adoption of its domestic propulsion systems in various marine applications. The cluster also mentions a separate report about Alibaba's potential $1.5 billion acquisition of grocery platform Pupu Supermarket and a review of a Nokia feature phone capable of running WeChat video. AI

Researchers have developed a new framework called GAT-MDN for more accurate salary prediction by considering the inherent uncertainty and multi-modal nature of compensation data. This approach utilizes Graph Attention Networks (GATs) to learn representations from job attributes like location and occupation, incorporating hierarchical and semantic relationships. The model then employs a Mixture Density Network (MDN) to output a full conditional salary distribution, outperforming traditional methods in experiments on a large Dutch job dataset. AI

A new research paper proposes a method called "signed compression progress" as a more robust form of intrinsic motivation for AI agents. This approach aims to ensure that an agent's reward is directly tied to genuine learning and improvement, rather than exploitable metrics. The paper provides a formal proof and experimental evidence demonstrating that this method resists common failure modes like reward clipping and exploitation of easily predictable outcomes. AI

SpaceX's recent IPO has reshaped the landscape of top venture capital investments, with its returns now dominating the historical top ten. Investments in SpaceX by firms like Valor Equity Partners and Founders Fund have yielded tens of billions of dollars, surpassing previously celebrated deals. This shift highlights the immense financial success of SpaceX's venture capital backing, with SoftBank and Naspers also featuring prominently for their investments in Alibaba and Tencent, respectively. AI

NATO's top military officer, U.S. Gen. Alex Grynkewich, is reassessing Europe's defense strategy due to anticipated U.S. cutbacks in aircraft and warships. The Pentagon is shifting focus to the Indo-Pacific, prompting calls for European allies and Canada to fill the resulting capability gaps, particularly in long-range fires and drones. These changes are being discussed ahead of a NATO summit in July, while a separate, smaller troop reduction is planned for NATO's Kosovo force. AI

A new technology called in-ovo sexing is being adopted in the U.S. egg industry to prevent the culling of male chicks. This method determines the sex of the chick while it is still in the egg, allowing for the separation of males before they hatch. Companies like Hidden Villa Ranch's NestFresh brand have fully transitioned to using this technology, with other farming groups also integrating in-ovo sexing machines into their hatcheries. Various technological approaches, including AI-powered imaging and DNA analysis, are being developed and refined for accuracy, speed, and cost-effectiveness. AI

SpaceX has officially set its share price at $135, aiming to raise $75 billion in what is poised to be the largest Initial Public Offering (IPO) in history. The company, led by Elon Musk, will trade on the Nasdaq under the ticker symbol SPCX. This valuation could potentially make Musk the world's first trillionaire, though questions remain about how SpaceX will justify its high valuation given its ambitious engineering projects, including a new American chip fabrication plant. AI

Researchers have developed a new framework to evaluate the credibility of simulated rainfall in autonomous driving perception tests. The method uses a path-based approach, representing each simulated path with equivalent rainfall intensity, an uncertainty band, and a realism score for raindrop distribution. This framework aims to better align simulated conditions with real-world rainfall, enabling more accurate testing and risk assessment for self-driving systems. AI

Researchers have developed a unified framework to understand when cross-modal alignment (CA) and cross-modal prediction (CP) are effective for multimodal learning. Their model identifies four distinct regimes: Both, CA only, CP only, and Neither, based on signal-to-noise ratios and cross-modal correlations. A data-driven procedure allows practitioners to diagnose their specific multimodal problem and select the appropriate objective before commencing training, potentially avoiding harmful cross-modal training in the 'Neither' regime. AI

The A-share index system is becoming more refined, with 374 new indices added this year, primarily focusing on technology themes like robotics, semiconductors, and artificial intelligence. Concurrently, pharmaceutical company WuXi AppTec has filed a lawsuit in a U.S. federal court against the Department of Defense. This action challenges the DoD's designation of WuXi AppTec as a Chinese Military Company (CMC) and its inclusion on the 1260H list, seeking to invalidate the decision and remove the company from the list. AI

China's State Council General Office has announced new guidelines to regulate "gambling agreements" (对赌协议) in private equity funds. This move aims to bring more clarity and oversight to these investment clauses, which have been a common practice but also a source of issues in the market. The new regulations are expected to shift the focus from strict contractual guarantees to the core competencies of investment institutions, such as market insight, post-investment support, and value discovery. AI

A Dutch chip startup, PHIX, is aiming to establish an all-European semiconductor manufacturing process. They are collaborating with an unnamed American partner to achieve this goal. The initiative seeks to bolster European semiconductor production capabilities. AI

A new research paper explores the interpretability challenges of using generative AI models in scientific domains with established theories. The study focuses on the 'Walrus' foundation model for continuum dynamics, employing sparse autoencoders to analyze its internal mechanisms. Researchers found that while the model can reproduce known dynamics, its internal representations are not always consistent with established physics, leading to discrepancies in output. AI

Researchers have developed GraphGP, a GPU-accelerated algorithm designed to make Gaussian processes more scalable. This new method utilizes Vecchia's approximation to reduce the computational complexity from cubic to linear, enabling the handling of nearly a billion parameters. Key innovations include a novel bit-reversed k-d tree ordering for efficient neighbor searches and parallel processing, alongside a differentiable CUDA implementation that significantly outperforms existing JAX baselines in speed and memory usage. AI

Hugging Face has developed a benchmark to evaluate how well automatic speech recognition (ASR) systems handle code-switched speech, where individuals switch between languages mid-sentence. This is crucial for voice agents serving bilingual customer bases. The benchmark, covering language pairs like Spanish-English and French-English, uses HR and IT service management scenarios. Top-performing models identified include ElevenLabs Scribe V2, Gemini 3 Flash, and Assembly AI Universal 3-Pro, with results reported using Word Error Rate (WER), Semantic Word Error Rate (SWER), and Answer Error Rate (AER). AI

A new research paper questions the effectiveness of Transformer models in network intrusion detection, particularly on the CIC-IDS2017 dataset. The study found that evaluation methodology, specifically padding conventions and data splitting, significantly impacts reported performance, often overestimating the Transformer's capabilities. When evaluated under realistic, leakage-free conditions without padding, the Transformer's performance drops considerably, suggesting that architectural choices are less critical than rigorous evaluation practices. AI

Researchers have developed deep learning models, specifically CNNs and GRUs, to differentiate between asthma and COPD using pulmonary sound data. The study optimized input representations like MFCC matrices and log-mel spectrograms, finding MFCCs to be superior. Adaptive-length windowing was crucial for handling inconsistent temporal dimensions in spectrograms, leading to the best cycle-based F1-score of 0.877 and subject-based F1-score of 0.855. AI

SpaceX has successfully launched its Initial Public Offering (IPO) on the Nasdaq, with its shares opening at $174, a 29% increase from its IPO price of $135. The company's stock began trading at $150 per share before rising to the opening price. This IPO is anticipated to make Elon Musk the world's first trillionaire. AI

Visa is integrating its payment network into OpenAI's ChatGPT to enable AI agents to make purchases on behalf of users. This collaboration aims to allow AI agents to shop for items like groceries or electronics and complete transactions across various merchants that accept Visa. The system will include security measures such as spending limits and approval steps to mitigate fraud and ensure consumer trust in AI-driven commerce. AI

Anthropic is launching a fellowship program called Claude Corps, which will place 1,000 AI coaches with nonprofits to help them utilize AI tools. This initiative, backed by a $150 million donation, aims to extend the benefits of AI while managing its risks. The company, valued at $965 billion and preparing for an IPO, emphasizes its commitment to balancing profit with social impact, a principle embedded in its public benefit corporation structure. AI

Anthropic has proposed a new economic framework to address AI-driven job displacement, including a potential tax on AI companies to fund support for affected workers. CEO Dario Amodei outlined policy recommendations, such as enhanced data collection, pro-employment incentives, and universal basic income, to ensure the benefits of AI are widely shared. The company is also launching a $200 million research fund and a $150 million fellowship program to study and mitigate the economic and societal impacts of artificial intelligence. AI

The Hugging Face Transformers library has released version 5.12.0, introducing new models like MiniMax-M3-VL, a vision-language model with a CLIP-style vision tower and a sparse Mixture-of-Experts decoder. This update also includes improvements to PP-OCRv6, an efficient OCR system, and Parakeet-RNNT, a fast conformer encoder with an RNN-T decoder. Additionally, version 5.11.0 added DiffusionGemma, an encoder-decoder model for faster text generation, and DeepSeek-V3.2-Exp, which features a novel sparse attention mechanism for long-context efficiency. AI

Researchers have developed NARRAS, a novel system for CSI-based localization in vehicular IoT networks. NARRAS employs an Edge-Triggered Distributed Inference (ETDI) approach, allowing remote antenna arrays to intelligently decide which channel state information (CSI) to report to a fusion center. This method optimizes resource usage by only transmitting valuable data, improving localization accuracy compared to other sparse-reporting strategies at similar uplink activity levels. AI

Researchers have developed a method to assess the image quality of identity cards for remote verification systems. This approach adapts quality measures from the Open Face Image Quality (OFIQ) standard to ID card images. The study found that applying these OFIQ measures can significantly enhance the performance of presentation attack detection algorithms. AI

Researchers have developed a new method called Manifold Power Iteration (MPI) to redesign the routers in Mixture-of-Experts (MoE) models. This technique aligns each router row with the principal singular direction of its associated expert, aiming to improve how tokens are routed to experts. Theoretical analysis suggests MPI drives router rows towards these principal directions, and empirical tests on MoE models ranging from 1B to 11B parameters show that this alignment leads to more effective models. AI

Researchers have developed Arbor, a novel AI framework designed for autonomous scientific research. Arbor utilizes a persistent knowledge tree called Hypothesis Tree Refinement (HTR) to link hypotheses, evidence, and insights, enabling cumulative learning across long-term projects. In evaluations across six research tasks, Arbor outperformed Codex and Claude Code, achieving over 2.5 times their average relative gain and reaching 86.36% Any Medal on MLE-Bench Lite with GPT-5.5. AI

Researchers have developed generalized conformal predictive systems (CPS) capable of handling distributional shifts in data. These systems encode shifts using observation-specific permutation weights, enabling them to produce calibrated predictive bands that adapt to varying data distributions. The approach introduces weight-uncertainty boxes to ensure confidence guarantees and has demonstrated effectiveness in experiments involving covariate shift and biomolecular design. AI

Researchers have developed GenAIR, a new framework designed to improve sequential recommendation systems by creating more effective item representations. This approach uses large language models to infer an "Archetype" for each item, representing its ideal target audience, and then grounds these archetypes in actual user behavior through a calibration objective. Experiments show that GenAIR significantly enhances the performance of various recommendation models across multiple datasets, outperforming existing methods. AI

President Donald Trump is reportedly nearing a deal to end a three-month war with Iran, coinciding with his 80th birthday and the return of the World Cup to the U.S. The agreement, if finalized, would aim to prevent Iran from developing nuclear weapons. This development follows escalating threats from Trump, including potential military action and seizure of Iranian oil facilities, which may have pressured Iran towards a settlement. AI

Poetic, an AI startup focused on enterprise automation, has secured $50 million in Series A funding at a $500 million valuation. The company, backed by OpenAI, Kleiner Perkins, Founders Fund, and First Harmonic, aims to revolutionize industries like healthcare and finance with its novel software approach. Poetic's technology, described as a new class of software that learns like AI but runs like code, promises higher accuracy and lower costs than traditional AI agents for mission-critical tasks. AI

Longevity biotech companies are attracting significant investment, raising $3.74 billion in Q1 2026, a 56% increase year-over-year. This surge is driven by AI's acceleration of biological research, the looming patent cliff for major drugs, and a shift towards treating aging as a root cause rather than just managing its symptoms. Companies like NewLimit and HexemBio are developing novel therapies, with a growing emphasis on clinical validation and regulatory progress. AI

Dinglong shares announced that its subsidiary, Qianjiang New Materials, has secured nearly 1,000 gallons of new orders for KrF/ArF photoresists from two leading wafer manufacturers. This development signifies a breakthrough in high-end photoresist products, which are critical consumables for logic and memory chip fabrication. The company has now had eight high-end wafer photoresist products receive bulk orders from major domestic wafer manufacturers, with five new additions since the end of the first quarter. AI

KKR, in partnership with the Kuwait Investment Authority, Nvidia, and Vestera, has launched Helix Digital Infrastructure. This new venture aims to provide essential infrastructure, including data centers, power, and connectivity, specifically for large cloud service providers focused on AI. Helix has secured over $10 billion in long-term funding commitments and will leverage Nvidia's AI factory infrastructure and Vestera's energy expertise. AI

Google has released DiffusionGemma, an experimental open-source model designed for faster text generation. This new model utilizes diffusion techniques to generate text blocks simultaneously, achieving speeds up to four times faster than previous Gemma models. The release aims to explore innovative approaches to text generation and is available under an Apache 2.0 license. AI

Researchers have developed a new Bayesian change-point detector called Span, designed to analyze serial circulating-tumour DNA (ctDNA) data. This method treats non-detects as left-censored observations, enabling the detection of drug resistance earlier than traditional methods. In simulations for metastatic breast cancer, Span approximately doubled the detection rate of impending progressions three months in advance compared to snapshot analyses. AI

Researchers have introduced TreeSeeker, a novel framework designed to improve the efficiency of deep search agents. This system structures search processes as a tree, allowing agents to explore multiple potential paths for complex queries while managing trial-and-error effectively. By employing a branch-and-return strategy and utilizing signals for value, uncertainty, and risk, TreeSeeker aims to prevent agents from getting stuck on unproductive paths and ensures better synthesis of evidence. Experiments demonstrate that TreeSeeker surpasses existing open-source methods in deep search tasks. AI

A new study published on arXiv evaluates the reliability of large language models (LLMs) in assessing the novelty of scientific research questions. Researchers developed a benchmark called RQ-Bench using recent arXiv papers to compare LLM-generated questions against author-anchored reference questions. The findings indicate that LLMs consistently overestimate the novelty of generated research questions, creating a "novelty mirage" that contradicts human expert evaluations. LLMs also tend to miss crucial dimensions like narrowness or source-binding in generated questions, raising concerns about their use in scientific evaluation. AI

Researchers have developed a new framework for training video diffusion models to solve general tasks by combining self-distillation and reinforcement learning. This method allows the models to learn task-solving abilities from unlabeled data, bypassing the need for costly, curated task-video supervision. The approach uses a vision-language model to generate tasks and solutions, which then guide a video diffusion model to learn execution, further enhanced by reinforcement learning from the vision-language model's feedback. AI

Researchers have introduced InternVideo3, a new framework designed to improve long-horizon video understanding and agentic capabilities. The system utilizes Multimodal Contextual Reasoning (MCR) to process video content as an evolving context, enabling evidence accumulation and verification over extended periods. To maintain efficiency, InternVideo3 incorporates Multimodal Multi-head Latent Attention (M^2LA), which compresses key-value cache states without losing token information. The model has demonstrated strong performance on various video understanding benchmarks and has been adapted into a video agent capable of evidence-grounded retrieval tasks. AI

Researchers have introduced "Structured PubMed," a large dataset containing over 23.2 million biomedical abstracts from PubMed. This dataset aims to improve information retrieval and text mining by providing section-labeled abstracts. It includes both author-structured abstracts and those automatically labeled using a Large Language Model pipeline, offering a valuable resource for training classification models and benchmarking text-segmentation architectures. AI

A new research paper introduces MASDR-RAG, a method to combat "vector search dilution" in retrieval-augmented generation (RAG) systems. This dilution occurs when scaling RAG to large document sets, leading to decreased accuracy as similarity searches return irrelevant information. The proposed solution involves scoping retrieval to specific domains using organizational metadata, which significantly improved performance in tests. AI

Researchers have introduced Annealed Entropic Allocation, a novel framework for sequential budget allocation in ranking and selection problems. This method employs an annealed weighted soft-min approach to refine the maximin objective, improving performance when multiple options are closely matched. The framework incorporates a saddlepoint approximation for enhanced discrimination with finite budgets, while maintaining the original large-deviation target as the smoothing parameter is annealed. AI

Researchers have developed AMNet, a novel multimodal framework for low-light video enhancement (LLVE) that can perform inference even when auxiliary data like infrared or event streams are unavailable. The system uses a Spatial-Spectral Dual-Gated Translator to generate implicit representations from RGB inputs, enabling robust enhancement. Extensive experiments show AMNet's superior performance in modality-absent conditions, with code and models publicly released. AI