Back to List
Meituan Technical Team Presents Breakthrough Research in Search and Recommendation at Top Global AI Conferences
Research BreakthroughMeituanArtificial IntelligenceLLM

Meituan Technical Team Presents Breakthrough Research in Search and Recommendation at Top Global AI Conferences

The Meituan Business R&D Platform's Search and Recommendation ASX (Agentic System X) team has recently highlighted its significant contributions to the field of Artificial Intelligence. Focusing on the development of Large Language Model (LLM)-based Agent technology systems, the team has achieved breakthroughs in LLM post-training, Agentic Reinforcement Learning, and Multi-modal understanding. Their research has been recognized by prestigious international conferences, including ICLR, NeurIPS, CVPR, and AAAI, with dozens of high-quality papers published. This article provides an overview of their research focus and highlights six selected papers that demonstrate Meituan's commitment to advancing Agentic systems and multi-modal AI capabilities within the search and recommendation landscape. The team's work underscores the growing importance of autonomous agents and sophisticated multi-modal processing in modern digital service platforms.

美团技术团队
Share

Key Takeaways

  • Strategic Focus on Agentic Systems: Meituan's ASX (Agentic System X) team is dedicated to building a comprehensive technology system centered on Large Language Model (LLM)-based Agents.
  • Core Research Pillars: The team's research is concentrated in three critical areas: LLM post-training, Agentic Reinforcement Learning, and Multi-modal understanding.
  • Global Academic Recognition: Meituan has published dozens of high-quality research papers in top-tier AI conferences, including ICLR, NeurIPS, CVPR, and AAAI.
  • Practical Application in Search and Recommendation: The research is specifically tailored to enhance the capabilities of search and recommendation systems within Meituan's Business R&D Platform.
  • Knowledge Sharing: The team has selected six representative papers for in-depth interpretation to provide insights and inspiration to the broader AI community.

In-Depth Analysis

The Strategic Vision of Agentic System X (ASX)

Meituan's Business R&D Platform has established the Search and Recommendation ASX (Agentic System X) team to spearhead the development of next-generation AI agents. This initiative reflects a broader industry shift from static AI models to dynamic, autonomous agents capable of complex reasoning and task execution. By focusing on an "Agentic System," Meituan aims to create a framework where Large Language Models (LLMs) serve as the core reasoning engine, enabling more interactive and intelligent search and recommendation experiences. The ASX team's focus on building this technology system suggests a long-term commitment to integrating agentic capabilities into the core of Meituan's digital services.

Core Research Directions: Post-training, RL, and Multi-modality

The technical depth of the ASX team is evidenced by their focus on three sophisticated research directions. First, LLM post-training is essential for refining general-purpose models to excel in specific domains, such as the nuances of local services and user intent in search. Second, Agentic Reinforcement Learning represents a frontier in AI where agents learn to make optimal decisions through interaction with their environment, a crucial component for personalized recommendation systems that must adapt to evolving user preferences. Finally, Multi-modal understanding allows the system to process and interpret diverse data types—including text, images, and potentially video—which is vital for a platform like Meituan where visual information (such as food photos or storefronts) is as important as textual descriptions.

Academic Excellence and Global Impact

The ASX team's success is validated by their consistent presence at the world's most prestigious AI conferences. Publishing dozens of papers in venues such as the International Conference on Learning Representations (ICLR), the Conference on Neural Information Processing Systems (NeurIPS), the Conference on Computer Vision and Pattern Recognition (CVPR), and the Association for the Advancement of Artificial Intelligence (AAAI) places Meituan at the forefront of global AI research. These conferences are highly competitive and serve as the primary stages for unveiling breakthroughs in machine learning and computer vision. By selecting six specific papers for interpretation, the team is not only showcasing their achievements but also contributing to the collective knowledge of the AI industry, particularly in how these advanced technologies can be applied to real-world search and recommendation challenges.

Industry Impact

The research conducted by Meituan's ASX team has significant implications for the AI industry, particularly in the evolution of search and recommendation engines. By successfully integrating LLM-based agents with reinforcement learning and multi-modal capabilities, Meituan is setting a benchmark for how large-scale platforms can leverage generative AI beyond simple chatbots. This work signals a move toward more proactive, context-aware systems that can understand user needs across different media and learn from complex interactions. Furthermore, the team's academic contributions help bridge the gap between theoretical AI research and practical, large-scale industrial application, influencing how other tech giants approach the development of agentic systems.

Frequently Asked Questions

Question: What is the primary focus of Meituan's ASX team?

The ASX (Agentic System X) team focuses on building a technology system based on Large Language Model (LLM) agents, specifically targeting improvements in search and recommendation functionalities through advanced AI techniques.

Question: In which research areas has the ASX team made significant contributions?

The team has deep expertise and has published research in three core areas: LLM post-training, Agentic Reinforcement Learning, and Multi-modal understanding.

Question: Which international conferences have recognized Meituan's research?

Meituan's research has been published in several top-tier AI conferences, including ICLR, NeurIPS, CVPR, and AAAI, totaling dozens of high-quality papers.

Read Original Article

Related News

Meituan LongCat Team Open-Sources WBench: A Systematic Multi-Round Evaluation Benchmark for Interactive Video World Models
Research Breakthrough

Meituan LongCat Team Open-Sources WBench: A Systematic Multi-Round Evaluation Benchmark for Interactive Video World Models

The Meituan LongCat technology team has announced the release and open-sourcing of WBench, the first systematic multi-round evaluation benchmark specifically designed for interactive video world models. As the industry shifts from passive video generation to active, interactive environments, WBench serves as a critical diagnostic tool—described by the team as a "CT scanner"—to identify exactly where current models struggle. By evaluating performance across diverse scenarios ranging from lunar walks to cybernetic cities, WBench aims to pinpoint the technical bottlenecks that prevent world models from achieving seamless interaction. This open-source initiative provides a structured framework for the AI community to measure and improve the interactive capabilities of next-generation world models, moving beyond simple observation to complex, multi-stage engagement.

Meituan Fulfillment AI Team Showcases Frontier Agent Technology and Research at ACL 2026 Conference
Research Breakthrough

Meituan Fulfillment AI Team Showcases Frontier Agent Technology and Research at ACL 2026 Conference

The Meituan Fulfillment AI Algorithm Team has recently highlighted its latest research achievements and technical practices featured at the ACL 2026 conference. Centered on developing a Large Language Model (LLM)-based Agent technology system, the team aims to revolutionize Meituan's fulfillment business through self-evolving operational systems. Their research focuses on critical AI frontiers, including Continuous Pre-training (CPT), Post-training, Agentic Reinforcement Learning (RL), and multimodal understanding. With a track record of dozens of high-quality papers published in prestigious international conferences such as ACL and EMNLP, Meituan's technical team continues to demonstrate its leadership in applying advanced AI agents to complex, real-world operational challenges in the fulfillment and delivery sector.

Google Research Explores the Algorithmic Foundations and Creativity of Diffusion Models
Research Breakthrough

Google Research Explores the Algorithmic Foundations and Creativity of Diffusion Models

Google Research has released a new publication titled "Towards demystifying the creativity of diffusion models," categorized under the domain of Algorithms & Theory. This research initiative focuses on providing a deeper, more theoretical understanding of how diffusion models—a cornerstone of modern generative AI—achieve creative outputs. By situating the study within algorithmic theory, Google Research aims to move beyond empirical observations of AI performance toward a robust mathematical framework. The goal is to demystify the complex processes that allow these models to generate novel and high-quality content, bridging the gap between technical execution and the perceived creativity of artificial intelligence. This work represents a significant step in the ongoing effort to understand the internal logic of generative systems.