The AI that solves the market: LG AI Research presents a new era in financial forecasting with natural language explainability
LG AI Research was established in 2020, as an offshoot of one of Korea’s largest multinationals with the goal of strengthening LG Group’s AI capabilities by addressing business challenges and pushing the cutting-edge of research and technology. Since then, the group has grown to comprise 250 AI scientists in support of LG’s various subsidiaries and affiliates, including LG Electronics, LG Energy Solutions, and many more.
The group’s AI efforts have been anchored around the development of EXAONE, which stands for Expert AI for Everyone. The foundation model was first released in 2021 as the first bilingual and multimodal generative AI model in Korea, and the team has continued to make progress in terms of model upgrades and commercialization results every year, culminating in the release of EXAONE Deep in 20225, a strong reasoning model with enhanced inference performance.
For Director of AI Business Development and Strategic Partnerships Young Choi, who specifically focuses on financial services and forecasting, the potential for EXAONE Deep and its companion models EXAONE 3.5 and EXAONE 4.0 to deliver value in financial forecasting was immense.
“Historically, a lot of forecasting in finance was just done using structured data, like numbers,” said Choi. “What we’re seeing with the entrance of LLMs is there’s a lot of leverage using text data like news reports, filings, transcripts. Now that we have context from news and text, you’re also able to explain the decision-making process, which can potentially solve the Black Box issue in financial services.”
Today, EXAONE powers a sophisticated and comprehensive AI framework that is powering the world’s first fully AI-managed ETF listed on the NYSE. Read on to learn more about how LG AI Research built Korea’s most sophisticated financial forecasting system and leveraged Weights & Biases to achieve state-of-the-art performance model performance.
The complex challenge of modern financial forecasting
Financial forecasting in today’s market presents challenges that traditional methodologies simply can’t handle. Global markets face immense complexity anticipating market shifts and demand fluctuations, while staggering data volumes create bottlenecks in translating information into predictive intelligence.
For LG AI Research, serving affiliate companies like LG Electronics, LG Energy Solution, and LG Household & Healthcare meant tackling forecasting problems with real consequences—missed opportunities and inefficient resource allocation measured in millions of dollars.
The team needed to overcome several critical challenges:
Data overload beyond traditional limits: Modern forecasting requires processing not just structured numerical data, but unstructured information from news, corporate filings, social media, and market reports—all in real-time.
The explainability problem: Financial services demand transparency in AI decision-making, but traditional models operate as black boxes, making it impossible to understand the reasoning behind predictions.
Multi-scale complexity: From microsecond-level market quotes to long-term strategic planning, the system needed to handle vastly different time horizons and data frequencies.
Real-world performance pressure: Unlike academic research, every prediction directly impacts multi-million dollar business decisions across LG’s diverse portfolio.
Building EXAONE and applying it to multiple domain use cases
At the core of LG AI Research’s forecasting capabilities lies EXAONE 3.5, their proprietary 32-billion parameter large language model that excels in long-context understanding and instruction following. The team also developed EXAONE Deep, a high-performance reasoning model that achieved the highest performance among Korean-developed models, scoring 83.0 on the MMLU benchmark.
“The EXAONE models were developed utilizing Weights & Biases to track training loss, validation, and evaluation throughout development,” said Choi. “Efficient management of learning trajectories is crucial for optimal performance, and we were able to accelerate improvements and achieve state-of-the-art performance within just a month – W&B was an integral part of EXAONE Deep’s success.”
The impact of LG AI Research’s forecasting models were immediate and measurable across several business units:
Supply and demand optimization: Forecasting models for LG Electronics and LG Household & Healthcare saved approximately $6.8 million annually by optimizing inventory management and production planning
Strategic raw materials forecasting: For LG Energy Solution, the team built models predicting lithium-ion battery prices to guide purchasing decisions for major clients including Tesla and Hyundai, enabling more strategic procurement timing.
Naphtha processing optimization: By managing the entire naphtha (a flammable liquid hydrocarbon mixture) processing chain—from raw material receipt to processing schedules—the system increased profits by approximately 13 million KRW per hour.
But perhaps the most ambitious – and measurable – application has been in public markets, through their LQAI ETF listed on the New York Stock Exchange.
LQAI: The world’s first fully AI-managed ETF and the EXAONE-GanphA agentic framework supporting it
The LQAI ETF represents a breakthrough in AI-driven investment management. Rebalancing every four weeks, the fund uses AI to select 100 stocks from the S&P 500, with decisions made entirely through their EXAONE-GanphA framework.
The system processes over 5,000 news articles daily, training continuously on market information to generate investment signals. The results speak for themselves: LQAI delivered 27.82% performance compared to SPY’s 24.89% and Goldman Sachs Active Beta Large Cap ETF’s 24.21%. The ETF also has a monthly generated explanatory report for the decisions it makes.
“Think of it as Warren Buffet’s annual letter, but generated entirely by AI,” said Choi. “Having the LLM generate explanations and reasoning for forecast results in natural language enhances understanding and trust in the AI decision making.”
The technical architecture behind these impressive results and outputs is EXAONE-GanphA, a modular AI framework designed to model, explain, and act within dynamic environments. The system employs four complementary AI agents working in concert:
AI Journalist: Curates timely market narratives by processing thousands of news sources daily, identifying relevant themes and sentiment patterns that traditional quantitative models miss.
AI Economist: Forecasts markets with full context, producing forward-looking signals by combining time-series modeling with exogenous conditioning and dynamic variable interaction analysis.
AI Analyst: A generative-explanatory module that translates numerical outputs into structured narratives, combining multi-source data fusion with scenario modeling and LLM-based explanation generation.
AI Decision-Maker: Turns insights into executable decisions, operationalizing insights through algorithmic ranking, optimization, and feedback simulation—whether for ETF rebalancing, supply chain optimization, or strategic planning.
This architecture is supported by three key technical innovations that LG AI Research developed through their forecasting projects:
Multi-modal deep document understanding (DDU): Transforms unstructured documents into formats that AI can comprehend and utilize, enabling the system to process corporate filings, research reports, and regulatory documents alongside traditional market data.
Advanced time-series analysis: Uses deep learning models like Transformers with mixed frequency, multi-variate, and ensemble learning methods to capture complex temporal patterns.
LLM-powered explainability: The same language model that processes market information generates explanations and reasoning for forecast results in natural language, solving the black box problem that has limited AI adoption in financial services.
Throughout the development of both EXAONE and the forecasting applications, Weights & Biases has been integral to LG AI Research’s success.
“We leverage heavily the product suite of Weights & Biases to build our models and fine-tune them,” said Choi. “There are many fine-tuning needs across all our affiliate companies, and we’re using it very heavily right now and are very happy with the product.”
The future of AI in financial services
LG AI Research’s work represents a fundamental shift in how AI can be applied to financial forecasting. By combining the power of large language models with traditional quantitative methods, they’ve created a system that doesn’t just predict—it explains, adapting to new information while providing the transparency that financial institutions require.
“We’re really excited to be pioneers in this space and lead the way,” said Choi. “Now that we have context from news and text, you’re also able to explain the decision making process and what the rationale is behind the outputs that AI’s are generating, which can potentially solve the Black Box issue which exists in financial services.”
As LG AI Research continues expanding their capabilities—from molecular structure prediction with EXAONE Discovery to pathological image analysis with EXAONE Path—their financial forecasting framework represents just the beginning of what’s possible when cutting-edge AI research meets real-world business challenges.
Learn more about the LQAI ETF here.