Meituan Highlights 12 Main-Conference Papers at EMNLP 2026
Meituan Technology Team said more than 20 of its papers were accepted by EMNLP 2026 and presented 12 main-conference long papers. The selection covers streaming voice agents, reinforcement learning, instruction-following evaluation, agent training, text-to-SQL, memory, tool-use data synthesis, neural interpretability, reward modeling, and long-context architectures. The descriptions and experimental results come from the team’s own publication, and the acceptance claims have not been independently confirmed in the supplied material.
The collection provides concrete methods and experimental findings across several active AI research areas, including the LTS-Agent framework for streaming voice interaction, although the evidence is company-authored.
EMNLP 2026 是国际计算语言学协会(Association for Computational Linguistics, ACL)举办的自然语言处理经验方法会议,是自然语言处理与计算语言学领域的国际顶级学术会议之一。
2026 年,美团技术团队共计 20 余篇论文被 EMNLP 2026 收录,我们从中精选了 12 篇主会论文(Long Paper)分享给大家。这些论文覆盖推理框架、策略优化、评估框架、强化学习、记忆框架、自适应训练框架、语义标签、数据合成框架、统计框架、奖励模型等技术领域和方向,希望能够对从事相关工作的同学以启发或者帮助。
1. A Listen-Think-Speak Framework for Efficient Streaming Voice Interaction via Semantic Triggering and Incremental Reasoning
- 论文类型:Main Conference,Long Paper
- 论文下载:PDF
#Meituan#LTS-Agent#Latent-GRPO#FlexBench#FlexEval#SIRI#SCOPE#TRUST-SQL#UniMem#ToolForge#RACE#F/S-RM