Allianz's Zeng on Fed Rate, De-Dollarization
Jenny Zeng, Allianz Global Investors Deputy Head of Fixed Income & APAC Fixed Income CIO, speaks on Bloomberg TV about the Fed rate and the trend of de-dollarization. (Source: Bloomberg)
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Yahoo
38 minutes ago
- Yahoo
Skywork-Reward-V2: Leading the New Milestone for Open-Source Reward Models
SINGAPORE, July 5, 2025 /PRNewswire/ -- In September 2024, Skywork first open-sourced the Skywork-Reward series models and related datasets. Over the past nine months, these models and data have been widely adopted by the open-source community for research and practice, with over 750,000 cumulative downloads on the HuggingFace platform, helping multiple frontier models achieve excellent results in authoritative evaluations such as RewardBench. On July 4, 2025, Skywork continues to open-source the second-generation reward models - the Skywork-Reward-V2 series, comprising 8 reward models based on different base models of varying sizes, with parameters ranging from 600 million to 8 billion. These models have achieved top rankings across seven major mainstream reward model evaluation benchmarks. Skywork-Reward-V2 Download Links HuggingFace: GitHub: Technical Report: Reward models play a crucial role in the Reinforcement Learning from Human Feedback (RLHF) process. In developing this new generation of reward models, we constructed a hybrid dataset called Skywork-SynPref-40M, containing a total of 40 million preference pairs. To achieve large-scale, efficient data screening and filtering, Skywork specially designed a two-stage human-machine collaborative process that combines high-quality human annotation with the scalable processing capabilities of models. In this process, humans provide rigorously verified high-quality annotations, while Large Language Models (LLMs) automatically organize and expand based on human guidance. Based on the above high-quality hybrid preference data, we developed the Skywork-Reward-V2 series, which demonstrates broad applicability and excellent performance across multiple capability dimensions, including general alignment with human preferences, objective correctness, safety, resistance to style bias, and best-of-N scaling capability. Experimental validation shows that this series of models achieved the best performance on seven mainstream reward model evaluation benchmarks. 01 Skywork-SynPref-40M: Human-Machine Collaboration for Million-Scale Human Preference Data Screening Even the most advanced current open-source reward models still perform inadequately on most mainstream evaluation benchmarks. They fail to effectively capture the subtle and complex characteristics of human preferences, particularly when facing multi-dimensional, multi-level feedback. Additionally, many reward models tend to excel on specific benchmark tasks but struggle to transfer to new tasks or scenarios, exhibiting obvious "overfitting" phenomena. Although existing research has attempted to improve performance through optimizing objective functions, improving model architectures, and recently emerging Generative Reward Models, the overall effectiveness remains quite limited. We believe that the current fragility of reward models mainly stems from the limitations of existing preference datasets, which often have limited coverage, mechanical label generation methods, or lack rigorous quality control. Therefore, in developing the new generation of reward models, we not only continued the first generation's experience in data optimization but also introduced more diverse and larger-scale real human preference data, striving to improve data scale while maintaining data quality. Consequently, Skywork proposes Skywork-SynPref-40M - the largest preference hybrid dataset to date, containing a total of 40 million preference sample pairs. Its core innovation lies in a "human-machine collaboration, two-stage iteration" data selection pipeline. Stage 1: Human-Guided Small-Scale High-Quality Preference Construction The team first constructed an unverified initial preference pool and used Large Language Models (LLMs) to generate preference-related auxiliary attributes such as task type, objectivity, and controversy. Based on this, human annotators followed a strict verification protocol and used external tools and advanced LLMs to conduct detailed reviews of partial data, ultimately constructing a small-scale but high-quality "gold standard" dataset as the basis for subsequent data generation and model evaluation. Subsequently, we used preference labels from the gold standard data as guidance, combined with LLM large-scale generation of high-quality "silver standard" data, thus achieving data volume expansion. The team also conducted multiple rounds of iterative optimization: in each round, training reward models and identifying model weaknesses based on their performance on gold standard data; then retrieving similar samples and using multi-model consensus mechanisms for automatic annotation to further expand and enhance silver standard data. This human-machine collaborative closed-loop process continues iteratively, effectively improving the reward model's understanding and discrimination of preferences. Stage 2: Fully Automated Large-Scale Preference Data Expansion After obtaining preliminary high-quality models, the second stage turns to automated large-scale data expansion. This stage no longer relies on manual review but uses trained reward models to perform consistency filtering: If a sample's label is inconsistent with the current optimal model's prediction, or if the model's confidence is low, LLMs are called to automatically re-annotate; If the sample label is consistent with the "gold model" (i.e., a model trained only on human data) prediction and receives support from the current model or LLM, it can directly pass screening. Through this mechanism, the team successfully screened 26 million selected data points from the original 40 million samples, achieving a good balance between preference data scale and quality while greatly reducing the human annotation burden. 02 Skywork-Reward-V2: Matching Large Model Performance with Small Model Size Compared to the previous generation Skywork-Reward, Skywork newly released Skywork-Reward-V2 series provides 8 reward models trained based on Qwen3 and LLaMA3 series models, with parameter scales covering from 600 million to 8 billion. On seven mainstream reward model evaluation benchmarks including Reward Bench v1/v2, PPE Preference & Correctness, RMB, RM-Bench, and JudgeBench, the Skywork-Reward-V2 series comprehensively achieved current state-of-the-art (SOTA) levels. Compensating for Model Scale Limitations with Data Quality and Richness Even the smallest model, Skywork-Reward-V2-Qwen3-0.6B, achieves overall performance nearly matching the previous generation's strongest model, Skywork-Reward-Gemma-2-27B-v0.2, on average. The largest scale model, Skywork-Reward-V2-Llama-3.1-8B, achieved comprehensive superiority across all mainstream benchmark tests, becoming the currently best-performing open-source reward model overall. Broad Coverage of Multi-Dimensional Human Preference Capabilities Additionally, Skywork-Reward-V2 achieved leading results in multiple advanced capability evaluations, including Best-of-N (BoN) tasks, bias resistance capability testing (RM-Bench), complex instruction understanding, and truthfulness judgment (RewardBench v2), demonstrating excellent generalization ability and practicality. Highly Scalable Data Screening Process Significantly Improves Reward Model Performance Beyond excellent performance in evaluations, the team also found that in the "human-machine collaboration, two-stage iteration" data construction process, preference data that underwent careful screening and filtering could continuously and effectively improve reward models' overall performance through multiple iterative training rounds, especially showing remarkable performance in the second stage's fully automated data expansion. In contrast, blindly expanding raw data not only fails to improve initial performance but may introduce noise and negative effects. To further validate the critical role of data quality, we conducted experiments on a subset of 16 million data points from an early version. Results showed that training an 8B-scale model using only 1.8% (about 290,000) of the high-quality data already exceeded the performance of current 70B-level SOTA reward models. This result again confirms that the Skywork-SynPref dataset not only leads in scale but also has significant advantages in data quality. 03 Welcoming a New Milestone for Open-Source Reward Models: Helping Build Future AI Infrastructure In this research work on the second-generation reward model Skywork-Reward-V2, the team proposed Skywork-SynPref-40M, a hybrid dataset containing 40 million preference pairs (with 26 million carefully screened pairs), and Skywork-Reward-V2, a series of eight reward models with state-of-the-art performance designed for broad task applicability. We believe this research work and the continued iteration of reward models will help advance the development of open-source reward models and more broadly promote progress in Reinforcement Learning from Human Feedback (RLHF) research. This represents an important step forward for the field and can further accelerate the prosperity of the open-source community. The Skywork-Reward-V2 series models focus on research into scaling preference data. In the future, the team's research scope will gradually expand to other areas that have not been fully explored, such as alternative training techniques and modeling objectives. Meanwhile, considering recent development trends in the field - reward models and reward shaping mechanisms have become core components in today's large-scale language model training pipelines, applicable not only to RLHF based on human preference learning and behavior guidance, but also to RLVR including mathematics, programming, or general reasoning tasks, as well as agent-based learning scenarios. Therefore, we envision that reward models, or more broadly, unified reward systems, are poised to form the core of AI infrastructure in the future. They will no longer merely serve as evaluators of behavior or correctness, but will become the "compass" for intelligent systems navigating complex environments, helping them align with human values and continuously evolve toward more meaningful goals. Additionally, Skywork released the world's first deep research AI workspace agents in May, which you can experience by visiting: Media Contact Company Name: Skywork AI Person: Peter TianEmail: peter@ 2 Science Park DriveCountry: SingaporeWebsite: View original content to download multimedia: SOURCE Skywork AI pte ltd Sign in to access your portfolio
Yahoo
an hour ago
- Yahoo
Apple Inc. (AAPL): People Are Tired Of The Stock Buybacks, Says Jim Cramer
We recently published . Apple Inc. (NASDAQ:AAPL) is one of the stocks Jim Cramer recently discussed. Cramer continues to be one of Apple Inc. (NASDAQ:AAPL)'s strongest proponents even though the firm's shares have lost 12.6% year-to-date. The stock has struggled due to trade tensions between the US and China that have threatened to disrupt the supply chain, the firm's struggle to convince the market about its presence in the AI market, and concerns about slow iPhone sales. However, the CNBC host believes that Apple Inc. (NASDAQ:AAPL) will maintain its stature as long as the firm holds its high-end smartphone market share. This time around, he criticized Apple Inc. (NASDAQ:AAPL)'s stock buybacks and deemed them inadequate: '[On reports of Apple reportedly looking to rely on third party AI] Look at how the stock reacted. Because people are tired of Apple just saying, you know what we're gonna do, we're gonna buy back stock until we get something better. No. I mean that's not what you can do anymore. A wide view of an Apple store, showing the range of products the company offers. Cramer commented on Apple Inc. (NASDAQ:AAPL)'s woes in detail recently. Here is what he said: '. . .Apple, which cannot get out of its own way. And I think probably could go down to 25 times earnings. Which is a substantial decline. Apple's a share donor. It's a share donor. '[On why Apple stock should be bought] No I'm not going to because I think the multiple's too high. While we acknowledge the potential of AAPL as an investment, our conviction lies in the belief that some AI stocks hold greater promise for delivering higher returns and have limited downside risk. If you are looking for an extremely cheap AI stock that is also a major beneficiary of Trump tariffs and onshoring, see our free report on the best short-term AI stock. READ NEXT: 20 Best AI Stocks To Buy Now and 30 Best Stocks to Buy Now According to Billionaires. Disclosure: None. This article is originally published at Insider Monkey.
Yahoo
2 hours ago
- Yahoo
Better Artificial Intelligence Stock: Nvidia vs. Meta Platforms
Nvidia has become the go-to provider for chips to support AI, which has resulted in explosive sales and earnings growth for the tech giant. Meta is pouring billions of dollars into AI research and aims to be a leader in the field. 10 stocks we like better than Nvidia › Artificial intelligence (AI) stocks have proven to be big winners for investors lately -- particularly last year, when some leading players in the space delivered double- and even triple-digit percentage gains. Though these high-growth companies' share prices tumbled earlier this year due to concerns about President Donald Trump's tariff plans, investors have recently returned to this compelling story. Trump's trade talks and tentative agreements on the frameworks of deals with the U.K. and China have boosted optimism that his tariffs won't result in drastically higher costs for U.S. consumers or major earnings pressure on U.S. companies -- in contrast to the worst-case scenario that many had feared. As a result, investors feel more comfortable investing in companies that rely on a strong economic environment to thrive -- such as AI sector players. This means that many investors are once again asking themselves which AI players look like the best buys today. Nvidia (NASDAQ: NVDA) and Meta Platforms (NASDAQ: META) are both aiming to reshape the future with their aggressive AI plans. If you could only buy one, which would be the better AI bet now? Nvidia already has scored many AI victories. The company has built an empire of hardware and services that make it the go-to provider for any organization creating an AI platform or program. But the crown jewels of its portfolio are its graphics processing units (GPUs). It offers the top-performing parallel processors, and thanks to both its ecosystem and manufacturing lead, they're also by far the best-sellers in their class. With demand from cloud infrastructure giants and other tech sector players still outstripping supply, Nvidia has been growing its sales at double- and triple-digit percentage rates, and setting new revenue records quarter after quarter. In its fiscal 2025, which ended Jan. 26, Nvidia booked a 114% revenue gain to a record level of $130 billion. And the company isn't just growing its top line -- its net income surged by 145% to almost $73 billion as it continued to generate high levels of profitability on those sales. Nvidia's clients today rely heavily on its hardware to power their projects, as its GPUs are some of the best chips available for the training of large language models (LLMs), as well as for inferencing -- the technical term for when those trained models are used to process real data to solve actual problems or make predictions. And Nvidia is helping customers with so much more -- from the design of AI agents to the powering of autonomous vehicle systems and drug-discovery platforms. Nvidia also is innovating steadily to stay ahead of rivals. It recently shifted to an accelerated schedule that will have it releasing chips based on new and improved architectures every year; previously, it rolled out new architectures about once every two years. So this company is likely to keep playing a major role in the evolution of AI throughout its next chapters. You will know Meta best as an owner of social media apps, some of which you probably use every day -- its core "family of apps" includes Facebook, Messenger, WhatsApp, and Instagram. And the sales of advertising space across those platforms have provided billions of dollars in revenue and profits for the company. But today, Meta's big focus is on AI. The company has built its own LLM, Llama, and made it open source so that anyone can contribute to its development. The open-source model can result in the faster creation of a better-quality product -- and in this case, it could help Meta emerge as a leader in the field. The company has put its money where its mouth is: It plans as much as $72 billion in capital spending this year to boost its AI presence. And just recently, Meta has been hiring up a storm in its efforts to staff its newly launched Meta Superintelligence Labs. That business unit will work on foundation models like Llama as well as other AI research projects. In a memo to employees regarding the new AI unit, Meta CEO Mark Zuckerberg highlighted why it's well positioned to lead in AI development: "We have a strong business that supports building out significantly more compute than smaller labs. We have deeper experience building and growing products that reach billions of people," he said. Those points are true, and they could help Meta reach its goals -- and deliver big wins to investors over time. From a valuation perspective, you might choose Meta, as the stock is cheaper in relation to forward earnings estimates than Nvidia -- a condition that has generally been the case. But a closer look shows that while Nvidia's valuation is down since the start of the year, Meta's actually has climbed. With that in mind, Nvidia looks like a more appealing buying opportunity, especially considering the company's ongoing strong growth and its involvement in every area of AI development and application in real-world situations. Meta also could emerge as a major AI winner down the road, and the stock is still reasonably priced today in spite of its gains in valuation. But Nvidia remains the key player in this space -- and at today's valuation, it's the better buy. Before you buy stock in Nvidia, consider this: The Motley Fool Stock Advisor analyst team just identified what they believe are the for investors to buy now… and Nvidia wasn't one of them. The 10 stocks that made the cut could produce monster returns in the coming years. Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you'd have $699,558!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you'd have $976,677!* Now, it's worth noting Stock Advisor's total average return is 1,060% — a market-crushing outperformance compared to 180% for the S&P 500. Don't miss out on the latest top 10 list, available when you join . See the 10 stocks » *Stock Advisor returns as of June 30, 2025 Randi Zuckerberg, a former director of market development and spokeswoman for Facebook and sister to Meta Platforms CEO Mark Zuckerberg, is a member of The Motley Fool's board of directors. Adria Cimino has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Meta Platforms and Nvidia. The Motley Fool has a disclosure policy. Better Artificial Intelligence Stock: Nvidia vs. Meta Platforms was originally published by The Motley Fool Error in retrieving data Sign in to access your portfolio Error in retrieving data Error in retrieving data Error in retrieving data Error in retrieving data
