<aside> 💡 Hi! This is the Decision LM group at MMLab@SIGS🤗. We are seeking cooperators to work on pushing pre-trained large models (LMs) as decision agents for diverse fields😊. If you are interested in working with us, feel free to contact Zhi Wang or Duo Wu🤗.
Zhi Wang
Email: [[email protected]](<mailto:[email protected]>)
Homepage: [<http://zwang.inflexionlab.org>](<http://zwang.inflexionlab.org/>)
Duo Wu
Email: [[email protected]](<mailto:[email protected]>)
Homepage: [<https://duowuyms.github.io/>](<https://duowuyms.github.io/>)
</aside>
Internship Recruitment (实习生招生，暂时关闭)
With billions of parameters pre-trained on massive data to absorb extensive knowledge, LMs have demonstrated extraordinary capabilities particularly in a myriad of NLP tasks. What’s more, they also exhibit emergent abilities that were not explicitly programmed into them during pre-training, such as planning, pattern mining, problem solving and generalization to unseen conditions.
Inspired by the remarkable success of LMs, we are dedicated to expand the boundaries of their capabilities and explore their potential in various domains, particularly in complex decision-making problems. In pursuit of this vision, our research interests primarily center on designing algorithms and systems that effectively harness the capabilities of LMs to address complex decision-making and planning problems to benefit diverse fields. We refer to this area of research as Decision LLM. Specifically, our current research directions include:
We are also excited to explore the potential of Decision LMs in multimedia networking and systems, robotic embodiment and industrial intelligence. The framework of our current research directions is illustrated as follows.
Many networking tasks now employ deep learning (DL) to solve complex prediction and system optimization problems. However, current design philosophy of DL-based algorithms entails intensive engineering overhead due to the manual design of deep neural networks (DNNs) for different networking tasks. Besides, DNNs tend to achieve poor generalization performance on unseen data distributions/environments.
Motivated by the recent success of large language models (LLMs), for the first time, this work studies the LLM adaptation for networking to explore a more sustainable design philosophy. With the massive pre-trained knowledge and powerful inference ability, LLM can serve as the foundation model, and is expected to achieve “one model for all” with even better performance and stronger generalization for various tasks. In this paper, we present NetLLM, the first LLM adaptation framework that efficiently adapts LLMs to solve networking problems. NetLLM addresses many practical challenges in LLM adaptation, from how to process task-specific information with LLMs, to how to improve the efficiency of answer generation and acquiring domain knowledge for net working. Across three networking-related use cases- view port prediction (VP), adaptive bitrate streaming (ABR) and cluster job scheduling (CJS), we showcase the effectiveness of NetLLM in LLM adaptation for networking. Results show that the adapted LLM surpasses state-of-the-art algorithms by 10.1-36.6% for VP, 14.5-36.6% for ABR, 6.8-41.3% for CJS, and also achieves superior generalization performance.
Duo Wu, Xianda Wang, Yaqi Qiao, Zhi Wang, Junchen Jiang, Shuguang Cui, Fangxin Wang. NetLLM: Adapting Large Language Models for Networking. Accepted by ACM SIGCOMM 2024 [CCF-A] [paper][project]
Utilizing large language models (LLMs) for tool planning has emerged as a promising avenue for developing general AI systems, where LLMs automatically schedule external tools (e.g. vision models) to tackle complex tasks based on task descriptions. To push this paradigm to ward practical applications, it is crucial for LLMs to consider tool execution costs (e.g. execution time) for tool planning. Unfortunately, prior studies overlook the tool execution costs, leading to the generation of expensive plans of which the costs outweigh task performance. To fill this gap, we propose the Cost-Aware Tool Planning with LLMs (CATP-LLM) framework, which for the first time provides a coherent design to empower LLMs for cost-aware tool planning. Specifically, CATP-LLM incorporates a tool planning language to enhance the LLM to generate non-sequential plans of multiple branches for efficient concurrent tool execution and cost reduction. Moreover, it further designs a cost-aware offline reinforcement learning algorithm to fine tune the LLM to optimize the performance-cost trade-off in tool planning. In lack of public cost-related datasets, we further present OpenCATP, the first platform for cost-aware planning evaluation. Experiments on OpenCATP show that CATP-LLM outperforms GPT-4 even when using Llama2-7B as its backbone, with the average improvement of 28.2%-30.2% higher plan performance and 24.7%-45.8% lower costs even on the challenging planning tasks. The codes of CATP-LLM and OpenCATP will be publicly available.
