Elsevier and LG AI Research have developed chemistry-specific AI vision technology that extracts chemical information from images in scientific literature and patents. The technology is now integrated into Elsevier's Reaxys chemistry discovery platform. Chemical substances and reactions are often depicted as drawings, figures, and reaction schemes rather than searchable text, making them difficult to index at scale. The new AI model combines molecule detection, reaction-diagram parsing, and optical chemical structure recognition to extract this visual chemistry more quickly and accurately than previous methods. LG AI Research's technology can interpret bonds, atoms, stereochemistry, and spatial relationships in chemical drawings. According to the company's published benchmarking, it outperforms alternatives at extracting chemistry from full document pages. The extraction pipeline has been validated against existing Reaxys benchmarks to maintain accuracy whilst increasing scale.

LG AI Research presents Expert AI Solutions Across manufacturing, finance and science at LG AI Talk Concert 2026. 13 hours ago Going beyond simply building good AI models to solving long-standing challenges in industrial fields, this is the path LG AI Research has walked and its core mission moving forward. On September 14, LG AI Research hosted the "LG AI Talk Concert 2026" at the LG Sciencepark Convergence Hall in Magok, Seoul. During the event, the institute presented key real-world applications of its "Expert AI," showcasing technological milestones and value created across manufacturing, finance, and science since its founding. "Expert AI that understands even the 1% exception proves its value in the field" In his keynote speech, Woohyung Lim, Head of LG AI Research, highlighted the clear distinction between general-purpose AI and domain-specific AI, emphasising the necessity of developing Expert AI. He said numerous general-purpose AI models already exist and generate impressive answers, but industrial environments demand AI that can understand complex variables and judge even the 1% exception with precision. He said AI must demonstrate tangible performance to prove its value in real-world operations. Following the keynote, Honglak Lee, Head and Chief AI Scientist (CSAI) of LG AI Research, took the stage alongside George Cameron, founder of Artificial Analysis, an independent research institution that objectively analyses and evaluates the performance, speed, intelligence, and cost of global AI models. The two discussed fair evaluation standards for global frontier AI models and shared insights into the recent rapid progress of Korean AI models and strategies to close the global technology gap. Under the theme of the "Future of AI," they also discussed how evaluation methodologies need to evolve and which capabilities will matter most. Transforming Industrial Fields: Expert AI Solutions Across 3 Key Sectors In Manufacturing, the first Deep Dive session, "Structured Data AI Innovation: Achieving Manufacturing AX with EXAONE Tabular," was held in collaboration with LG Innotech. The team introduced domain-specific foundation models, including EXAONE Tabular, which predicts process conditions and quality, and EXAONE Omni-Inspect, designed to inspect products without re-training even when process conditions or visual appearances change. Through collaboration with LG Innotech, LG AI Research established a fast, accurate prediction system using minimal field data. By deploying a vision inspection agent that autonomously handles data sampling, labelling, and model training, the institute is accelerating the transition toward autonomous factories. In Finance, the second session, "Financial Data Innovation and Insights Driven by EXAONE Business Intelligence and Partnerships," brought together experts from the National Pension Service (NPS), KOSCOM, and the London Stock Exchange Group (LSEG) for in-depth discussions. Key accomplishments of EXAONE BI, a specialised financial solution delivering end-to-end data analysis, reasoning, prediction, and explanation, took centre stage. EXAONE BI is actively expanding its footprint by targeting global markets with LSEG and domestic markets with KOSCOM, while extending its coverage beyond North America, Europe, and the Middle East to encompass a wider range of asset classes. In Science, the third session, "Beyond Automation to Autonomy: The Future of Materials R&D Presented by AI Autonomous Labs," was conducted with representatives from LG Chem and GS Caltex. Together with the three experts, LG AI Research presented a vision for an autonomous laboratory ecosystem that goes beyond material design and synthesis prediction to directly perform physical experiments. Over 100 Industrial Challenges Solved: Moving Toward Robot Foundation Models Since its establishment, LG AI Research has solved over 100 complex industrial challenges, including battery lifespan and capacity prediction, defect detection, and supply chain optimisation. These achievements are backed by top-tier research capabilities, demonstrated by 368 papers accepted at leading global AI conferences and 1,080 patents filed worldwide. Looking ahead, LG AI Research shared its progress on developing Robot Foundation Models, enabling AI to perceive, decide, and act in physical environments. Prioritising safety through proprietary core algorithms, the institute aims to advance beyond individual robot automation to create autonomous factories where an entire facility operates under a unified intelligence. "EXpert AI for everyONE" LG AI Research will continue its journey to ensure the true value embedded in the name EXAONE reaches both industrial fields and everyday life.

LG AI discovers hair-loss treatment material in a day, eyes manufacturing next. LG's science-focused AI analyzed 420,000 substances to identify Rhamsydil, a potential female pattern hair-loss treatment for Dr.Groot products. Published September 14, 2026 - 5:14 p.m. LG's science-focused AI model has discovered a new hair-care material by analyzing 420,000 candidate substances in a single day. LG Household & Health Care, the group's cosmetics and consumer goods affiliate, plans to apply the material to its scalp-care brand Dr.Groot. The material, named Rhamsydil, helps ease female pattern hair loss by creating a scalp environment more conducive to hair growth. "We have selected industrial challenges based on where AI can have the greatest effect and scalability, and have driven innovative change as a result," said Lim Woo-hyung, co-head of LG AI Research, in a keynote address. "That approach has allowed us to solve more than 100 industrial problems to date." The model behind the discovery, Exaone Discovery, is tailored to the sciences, designing new materials and predicting synthesis outcomes. It is one of several models LG AI Research has built for specific industries. Exaone Tabular, aimed at manufacturing, reads and predicts process conditions and quality data. Exaone Omni-Inspect is an autonomous visual inspection model that adapts to changes in a product's or process's appearance without needing to be retrained each time. Exaone Business Intelligence, built for finance, uses multiple AI agents working together to handle data analysis, inference, forecasting and explanation, drawing on Exaone Forecast, a foundation model that tracks how information changes over time. Exaone 4.5 is the company's general-purpose large language model. LG AI Research has also begun developing a robot foundation model that would let robots judge and act in physical environments, a step toward autonomous factories. Working with affiliates including LG Innotek and LG Display, the research lab found that the harder challenge was not building a model or deploying it, but keeping it working afterward. "Manufacturing conditions change constantly," said Lee Hwa-young, head of AI business development at LG AI Research. "The problem shifts, operating conditions shift - the on-site situation is so dynamic in ways that break models trained on a fixed snapshot of data." LG's manufacturing models are designed to run on minimal GPU capacity, making it easier to deploy them on-site rather than through the cloud. A major source of hesitation around AI adoption in manufacturing is the risk of domain knowledge leaking overseas, since routing data through an external cloud provider raises that exposure. Running models on-premise, inside a factory's own infrastructure, avoids that risk while also cutting costs, since larger models require more GPUs and drive up expenses as they scale. BY LEE JAE-LIM [[email protected]]