en

Please fill in your name

Mobile phone format error

Please enter the telephone

Please enter your company name

Please enter your company email

Please enter the data requirement

Successful submission! Thank you for your support.

Format error, Please fill in again

Confirm

The data requirement cannot be less than 5 words and cannot be pure numbers

m.nexdata.datatang.com

Meet Nexdata at ICML 2026

From:Nexdata Date: 07/03/2026

Nexdata will take part in the 43rd International Conference on Machine Learning, ICML 2026, which will be held from July 6 to July 11, 2026, at the COEX Convention & Exhibition Center in Seoul, South Korea. Nexdata will welcome attendees at Booth B605.

At the conference, Nexdata will demonstrate its data solutions for AI model development across four main directions: GenAI/VLM, Physical AI, Speech LLM, and LLM. The company looks forward to connecting with AI researchers, enterprise customers, and industry partners from around the world to discuss the importance of data in model training, evaluation, optimization, and deployment.

Come to Meet Nexdata at ICML 2026

ICML is one of the most renowned conferences in machine learning, bringing together researchers, research institutions, technology companies, and industry innovators to exchange the latest results in machine learning theory, methods, systems, and applications.

As AI models continue to progress, their data requirements are also changing. AI teams are moving from single-modal and single-task datasets toward higher-quality, more scalable, and more task-oriented data foundations.

From Generative AI and multimodal models to speech language models, agent systems, and Physical AI, high-quality data is becoming increasingly important for building, validating, and optimizing next-generation AI models.

Four Main AI Data Solutions Presented by Nexdata

At ICML 2026, Nexdata will demonstrate its off-the-shelf data resources and data service capabilities across four main AI directions, helping AI teams build, evaluate, and optimize models more efficiently.

Gen-AI / VLM

For the development of generative AI, visual language models, and multimodal AI, Nexdata will demonstrate data capabilities for images, videos, image-text data, multimodal understanding, image editing, and GUI Agent scenarios. These data resources can support model training for content understanding, generation, interaction, and multimodal tasks.

Physical AI

For robot learning and physical-world interaction, Nexdata will demonstrate Physical AI data capabilities such as Ego-centric data, 3D scene data, and dexterous manipulation data. These solutions support robotics model training in environment perception, task understanding, operation execution, and scene interaction.

SpeechLLM

For the development of speech language models, dialogue models, and intelligent voice interaction systems, Nexdata will demonstrate data capabilities such as full-duplex dialogue, spontaneous dialogue, Conversational TTS, English speech data from different countries and regions, and multilingual speech data. These solutions support model development in speech understanding, speech generation, dialogue interaction, and cross-language applications.

LLM

For large language model training, reasoning, and evaluation, Nexdata will demonstrate data capabilities in code data, long-context QA and reasoning data, STEM test questions, Agent interaction data, and model evaluation. These solutions are designed to improve model capabilities in understanding, generation, reasoning, and complex task execution.

From Off-the-Shelf Data to Bespoke Services

In addition to off-the-shelf datasets, Nexdata will demonstrate its bespoke data service capabilities in data collection, data processing, data annotation, quality management, model evaluation, and benchmarking.

Based on different model types, application scenarios, and project stages, Nexdata provides flexible data solutions to help clients build a complete data support path from data preparation and model training to evaluation and optimization.

Since 2011, Nexdata has continuously provided high-quality AI data products and services for global AI companies, research institutions, and industry clients. Its service scope covers speech, images, text, multimodal AI, intelligent driving, generative AI, and Physical AI. With long-term project experience, scalable delivery capabilities, and a structured quality management system, Nexdata helps accelerate model innovation through better data.

Visit Nexdata at Booth B605

During ICML 2026, Nexdata invites attendees, AI researchers, enterprise customers, and industry partners to visit Booth B605 to learn more about its data solutions for GenAI/VLM, Physical AI, Speech LLM, and LLM, and to explore new opportunities in AI data development.

36704bd1-3d5c-4eaa-ae3c-9ae9782df3df