Researchers have leveraged Climate Data and Satellite Technology for developing the predictive model

CHANDIGARH, India, June 27, 2026 /PRNewswire/ — In a significant advancement for smart agriculture and precision farming, researchers at Chandigarh University have developed an Artificial Intelligence (AI)-powered Transformer Model that is capable of accurately predicting crop yields using satellite imagery, climate data and historical agricultural records.

The innovation would be instrumental in empowering farmers, policymakers and agricultural agencies to make informed decisions while strengthening food security and advancing resilient farm management.

The research, led by Kusum Lata, Assistant Professor Department of Computer Science Engineering, Chandigarh University, Dr Navneet Kaur Professor Department of CSE and Dr Simrandeep Singh Professor from University Centre of Research and Development at Chandigarh University that focuses on improving crop yield forecasting in Punjab’s agricultural heartland. The study, recently presented at the 2026 International Conference on Signal Processing and Electronics Design (ICSPED) at Chandigarh College of Engineering and Technology, Chandigarh that introduces a lightweight transformer-based system that leverages multi-source data to estimate crop production before harvest with greater accuracy and lower computational costs.

Notably, the accurate crop yield prediction has become increasingly important as farmers face growing challenges from climate variability, changing weather patterns and rising food demand. Traditional field surveys are often time-consuming, labour-intensive and limited in scale. The Chandigarh University researchers sought to overcome these limitations by integrating advanced AI techniques with real-time Earth observation data.

Kusum Lata, Assistant Professor Department of Computer Science Engineering at CU said, “The transformer model utilizes data from Sentinel-1 and Sentinel-2 satellites which are advanced Earth observation satellites operated by the European Space Agency (ESA) to continuously monitor agricultural fields and provide information on crop growth, vegetation health, soil moisture and field conditions. The satellite observations are combined with climatic variables such as rainfall, temperature and soil moisture, along with historical crop production records, creating a comprehensive picture of crop performance throughout the growing season.”

Kusum added, “Unlike conventional machine learning models, the newly developed lightweight transformer can identify critical crop growth stages and learn complex temporal patterns that influence final yields. We have designed the model to deliver high predictive performance that require fewer computational resources, making it suitable for practical deployment in large-scale agricultural monitoring systems.”

“The model was evaluated on four major crops cultivated in Ludhiana district, namely paddy, maize, moong and sugarcane, using data collected between 2019 and 2023. Experimental results demonstrated that the transformer model outperformed widely used approaches such as Random Forest and Long Short-Term Memory (LSTM) models, indicating stronger agreement between predicted and actual yields. The framework also recorded lower prediction errors and improved computational efficiency.

Kusum has worked as a Junior Research Fellow in the Agriculture and Crop Monitoring Division at Punjab Remote Sensing Centre (PRSC), PAU Ludhiana, contributed to geospatial research projects, including crop residue burning analysis using remote sensing and geospatial mapping techniques.

The study further revealed that the lightweight architecture requires nearly 40 percent fewer parameters than conventional transformer models while delivering faster and accurate predictions. This level of efficiency makes this automatic framework suitable for near real-time agricultural applications, including regional crop monitoring, production forecasting and early warning systems.

According to the researchers, the ability to accurately forecast crop yields before harvest can have significant implications for farmers and governments alike. Reliable forecasts can support agricultural planning, optimize resource allocation, strengthen crop insurance mechanisms and improve market management strategies. In a state like Punjab, where agriculture plays a central role in the economy, such technologies can contribute to more resilient and sustainable farming systems.

The researchers also shared that one of the key strengths of the model lies in its ability to combine multiple sources of information into a single predictive model. By integrating satellite-derived observations with climatic and historical datasets, the system captures the complex interactions that influence crop productivity and provides a more robust understanding of agricultural outcomes.

The future developments will also focus on enabling near real-time forecasting through cloud-based platforms, paving the way for broader adoption of AI-driven decision support systems in agriculture, added the Chandigarh University researchers.

About Chandigarh University

Chandigarh University is a NAAC A+ Grade University and QS World Ranked University. This autonomous educational institution is approved by UGC and is located near Chandigarh in the state of Punjab. It is the youngest university in India and the only private university in Punjab to be honoured with A+ Grade by NAAC (National Assessment and Accreditation Council). CU offers more than 109 UG and PG programs in the field of engineering, management, pharmacy, law, architecture, journalism, animation, hotel management, commerce, and others. It has been awarded as The University with Best Placements by WCRC.

Website address: https://www.cuchd.in/

Photo: https://mma.prnewswire.com/media/3001456/Chandigarh_University_AI_Crop_Yield.jpg

View original content to download multimedia:https://www.prnewswire.com/in/news-releases/chandigarh-university-researchers-use-artificial-intelligence-to-develop-model-for-predicting-accurate-crop-yield-innovation-to-benefit-indian-farmers-302812564.html

HONG KONG, June 27, 2026 /PRNewswire/ — Waton Financial Limited (NASDAQ: WTF) today announced the release of MoTA Alpha, a major upgrade to its flagship AI-powered investment platform. First unveiled in closed beta in May 2026, MoTA (Manager of Trading Agents) now introduces the Agent Talents Market, a creator ecosystem for third-party AI trading agents, alongside a redesigned multi-agent collaboration workflow and a fully overhauled user experience. The Alpha release represents more than a product milestone — it signals Waton’s decisive transition from a securities brokerage and SaaS provider into an AI-native financial technology company.

MoTA Alpha: What’s New

MoTA Alpha builds on the beta’s foundation as an AI-native investment team workbench — a platform that enables professional investors and portfolio managers to assemble, manage, and supervise teams of specialized AI agents across research, analysis, risk, and execution functions within a structured, auditable workflow, with mandatory human review and final sign-off at every stage.

Three headline upgrades define this release:

Agent Talents Market
An open marketplace where third-party developers can create, publish, and rank AI trading agents. Users subscribe to or deploy agents built by independent creators, with all agents running on Waton’s infrastructure. Agent’s internal logic remains under creator control; Waton provides the platform layer and execution environment.

Enhanced Multi-Agent Collaboration
A rebuilt task orchestration layer that improves inter-agent communication, role assignment, and decision audit trails. The result is a workflow that mirrors the dynamics of a real investment team — each agent operates within its mandate, escalates to human supervisors where required, and maintains a complete, reviewable log.

Redesigned Interface
A significantly improved user experience that preserves MoTA’s distinctive 8-bit pixel-art visual identity — a deliberate departure from the blue-and-white minimalism that dominates fintech — while increasing information density and operational speed for professional workflows.

The Strategic Pivot

MoTA Alpha represents the clearest demonstration yet of Waton’s evolution from a financial infrastructure provider into an AI-native product company.

Since its NASDAQ listing in April 2025, Waton has positioned itself as the world’s first publicly traded AI agent holding company. Yet its revenue base has remained rooted in traditional securities brokerage and Broker Cloud SaaS solutions serving institutional clients in Hong Kong. MoTA Alpha changes that equation: AI is no longer a narrative layer on top of an existing brokerage business — it is now a tangible, independently monetizable product line.

The company is structuring itself around a “brokerage infrastructure + AI application” dual-engine model. This is a meaningfully different profile from either pure-play online brokers or conventional fintech SaaS firms, positioning Waton closer to the emerging category of AI-native financial platforms.

Financial Foundation

According to Waton’s unaudited financial results for the first half of fiscal year 2026 (six months ended September 30, 2025), total revenues rose 106.3% year-on-year to $6.10 million, driven by a 223.1% increase in brokerage and commission income to $4.17 million. Cash and segregated cash stood at $29.88 million, with total assets of $68.98 million.

Notably, the company reported research and development expenses as a standalone line item for the first time ($0.39 million in H1 FY2026), alongside significant share-based compensation tied to AI product development. MoTA Alpha is the first scaled output of this R&D pipeline.

Management Commentary

“The Alpha release of MoTA marks Waton’s evolution from a financial technology services provider to an AI-era infrastructure and product company,” said Zhou Kai (Tony Zhou), Chairman and Chief Technology Officer of Waton Financial. “We are not building a chatbot for trading. We are building a platform where professional investors manage teams of AI agents — each with defined roles, clear accountability, and human oversight. The Agent Talents Market extends this further: MoTA transitions from a product into an ecosystem.”

Roadmap

Following the Alpha release, Waton expects to open MoTA to public beta testing in Q3 2026. The platform currently supports Hong Kong and U.S. equity markets, with digital asset coverage on the product roadmap. MoTA is available as a standalone application at m.mota.ai and integrates with Waton’s existing brokerage and TradingWTF infrastructure.

For investors tracking $WTF, MoTA Alpha serves as the first real test of whether the “AI agent holding company” thesis translates from corporate positioning into a durable commercial model.

Media Contact

https://www.wtf.us

https://www.mota.ai 

About Waton Financial Limited

Waton Financial Limited (NASDAQ: WTF) is the world’s first NASDAQ-listed AI agent holding company, dedicated to discovering, creating, investing in, and incubating AI agents that work for people. Its flagship product, MoTA (Manager of Trading Agent), enables professional investors to build, manage, and supervise teams of specialized AI agents within a structured, human-supervised workflow. The company also empowers global brokerage firms and financial institutions through Broker Cloud + SaaS + AI digital solutions. Learn more at https://wtf.us.

View original content:https://www.prnewswire.com/news-releases/waton-financial-launches-mota-alpha-marking-full-strategic-pivot-to-ai-native-finance-302812548.html

SOURCE Waton Financial Limited

HONG KONG, June 27, 2026 /PRNewswire/ — Chuangxinzhong, a precision marketing subsidiary of Yeahka (9923.HK), has become the top-ranked partner by model consumption among all agency-tier partners of ByteDance’s Jichuang 2.0 model.

According to Chuangxinzhong personnel, the company has deployed AI agents to automate manual processes since May, driving a 116% year-on-year increase in model usage. Meanwhile, its AI content production capacity rose by 33%, enabling the company to generate 30 to 40 video ad creative sets per day.

The figures mark a significant step forward for Chuangxinzhong in the commercialization of AIGC and the large-scale production of short-video ad creatives.

As short-video platforms such as Douyin continue to surge, traditional advertising models are facing mounting pressure. With consumer attention increasingly scarce, advertisers are widely grappling with rising costs and declining ROI. This is particularly true in fast-moving categories like beauty and apparel, where a single round of product testing often requires a dozen or more short-video assets, and the cycle from creative concept to launch can be lengthy.

For advertisers across the industry, ad placement has entered an era of “creative-driven growth”, especially in sectors such as finance, e-commerce, and local services, where a single creative concept is no longer enough to sustain growth — companies now need large volumes of high-quality content to test and optimize quickly.

Through its collaboration with the Jichuang 2.0 model, Chuangxinzhong’s monthly AI-generated ad spend rose from RMB 5 million to RMB 10 million, underscoring the high-frequency use of its AI content generation capabilities. Meanwhile, its daily output of 30 to 40 asset sets has directly translated into greater agility for advertisers responding to market shifts.

This is not the first breakthrough Chuangxinzhong has achieved in AI-driven marketing. The company previously ranked first in AIGC spend within the financial lead-generation sector during a digital human ad incentive competition hosted by ByteDance, where it also set an industry record: an 80% reduction in per-asset cost alongside a 391% week-on-week increase in consumption.

A representative from Chuangxinzhong said the company will continue investing in AI marketing infrastructure going forward, aiming to deepen the integration of LLM capabilities with advertising, user operations, and business growth. The goal is to offer enterprise clients an integrated solution spanning creative production, intelligent ad placement, and performance optimization.

As a key part of Yeahka’s broader AI strategy, Chuangxinzhong has spent recent years driving the adoption of AI technologies in marketing. Building on the group’s accumulated expertise in large models, algorithms, and content generation, the company has developed a product suite spanning digital humans, AIGC content production, and smart marketing.

From digital humans to AIGC content factories to intelligent marketing agents, AI is steadily reshaping how the marketing industry produces content. Chuangxinzhong’s top performance in ByteDance’s Jichuang 2.0 agency rankings reflects the strong, scalable capabilities and industry-leading position the company has built in commercial AI applications.

View original content:https://www.prnewswire.com/apac/news-releases/yeahkas-chuangxinzhong-tops-bytedances-jichuang-2-0-agency-rankings-signaling-acceleration-of-ai-content-production-302812515.html

SOURCE Yeahka