Auto-regressive generation models achieve competitive performance across many different NLP tasks such as summarization, question answering, and classifications. However, they are also known for being slow in inference, which makes them challenging to deploy in real-time applications. We propose a switchable decision to accelerate inference by dynamically assigning computation resources for each data instance. Automatically making decisions on where to skip and how to balance quality and computation cost with constrained optimization, our dynamic neural generation networks enforce the efficient inference path and determine the optimized trade-off. Experiments across question answering, summarization, and classification benchmarks show that our method benefits from less computation cost during inference while keeping the same accuracy. Extensive experiments and ablation studies demonstrate that our method can be general, effective, and beneficial for many NLP tasks.
A Graph-Retrieval-Augmented Generation Framework Enhances Decision-Making in the Circular Economy
Singapore Institute of Manufacturing Technology (SIMTech)(新加坡制造技术研究所);Agency for Science, Technology and Research (A*STAR)(科学、技术及研究局);The University of Sydney(悉尼大学);Nanyang Technological University(南洋理工大学);National University of Singapore(国立新加坡大学)
Bridging the Gap between Expert and Language Models: Concept-guided Chess Commentary Generation and Evaluation
Graduate School of AI, POSTECH(浦项科技大学人工智能研究生院);Department of Computer Science and Engineering, POSTECH(浦项科技大学计算机科学与工程系);Department of Chemical Engineering, POSTECH(浦项科技大学化学工程系);KRAFTON
Towards Feedback-to-Plan Decisions for Self-Evolving LLM Agents in CUDA Kernel Generation
Department of Computer Science and Technology, Tsinghua University, Beijing, China(清华大学计算机科学与技术系);Beijing National Research Center for Information Science and Technology, Beijing, China(北京信息科学国家研究中心)
EEG-MedRAG: Enhancing EEG-based Clinical Decision-Making via Hierarchical Hypergraph Retrieval-Augmented Generation
College of Information Science and Engineering, Northeastern University(信息科学与工程学院,东北大学);Foshan Graduate School of Innovation, Northeastern University(创新佛山研究生院,东北大学);College of Computing and Data Science, Nanyang Technological University(计算与数据科学学院,南洋理工大学);Institute of Automation, Chinese Academy of Sciences(自动化研究所,中国科学院);Squirrel Ai Learning
Understanding the Fundamental Design Decisions of Retrieval-Augmented Generation Systems
Fudan University(复旦大学);East China Normal University(华东师范大学);Shanghai Innovation Institute(上海创新研究院);The University of Tokyo(东京大学);Macau University of Science and Technology(澳门科学理工学院);Concordia University(Concordia大学);University of Alberta(阿尔伯塔大学);The University of Tokyo, Japan(日本东京大学)
Uncertainty-Aware Generation and Decision-Making Under Ambiguity
Ubiquitous Knowledge Processing Lab (UKP Lab), Department of Computer Science, Technical University of Darmstadt(普遍知识处理实验室(UKP实验室),计算机科学系,达姆施塔特技术大学);National Research Center for Applied Cybersecurity ATHENE, Germany(应用网络安全国家研究中心ATHENE,德国)
Generating One-Anaphoric Expressions: Where Does the Decision Lie?
大多数自然语言生成系统包含用于选择是否随后借助于代词或定名词短语来指代已经引入的实体的机制。然而,相对而言,很少有系统考虑通过一个照应的表达方式来指代实体,例如\lingform{the small green one}。本文着眼于什么是参与生成这种类型的指称表达式。考虑到如何适应这种能力到一个标准的算法,涉及表达式生成导致我们建议修改现有的方法的基础上的一些假设。我们证明了我们的方法的有用性,在一个简单的数据库接口应用程序的上下文中的一个照应生成,并提出了一些意见,这种方法的影响,指表达生成更普遍。