Skip to content
jinlong edited this page Jun 29, 2023 · 4 revisions

LLM

LLM+Tools

Fine-tuning LLMs to use tools.

  • Webgpt: Browser-assisted question-answering with human feedback. Preprint 2023 [pdf] [blog]

    Reiichiro Nakano, Jacob Hilton, Suchir Balaji, Jeff Wu, Long Ouyang, Christina Kim, Christopher Hesse, Shantanu Jain, Vineet Kosaraju, William Saunders, Xu Jiang, Karl Cobbe, Tyna Eloundou, Gretchen Krueger, Kevin Button, Matthew Knight, Benjamin Chess, John Schulman

    Annotation: fine-tune GPT-3 on human web search behaviors to learn how to use the web browser.

  • Talm: Tool augmented language models. Preprint 2022 [pdf]

    Aaron Parisi, Yao Zhao, Noah Fiedel

    Annotation: use pretrained T5 models [Raffel et al., 2019,Roberts et al., 2022] for finetuning, inference and evaluation. use the base, large, and XL sizes.

  • Toolformer: Language Models Can Teach Themselves to Use Tools. Preprint 2023 [pdf], [Code]

    Timo Schick, Jane Dwivedi-Yu, Roberto Dessì, Roberta Raileanu, Maria Lomeli, Luke Zettlemoyer, Nicola Cancedda, Thomas Scialom

    Annotation: fine-tune on GPT-J

  • ToolkenGPT: Augmenting Frozen Language Models with Massive Tools via Tool Embeddings Preprint 2023. [pdf]

    Shibo Hao, Tianyang Liu, Zhen Wang, Zhiting Hu

    Annotation: learns an embedding tool as a token (“toolken”) in llama 30B

In-context learning for tools.

  • Program of thoughts prompting: Disentangling computation from reasoning for numerical reasoning tasks. Preprint 2022 [pdf]

    Wenhu Chen, Xueguang Ma, Xinyi Wang, William W. Cohen

    Annotation: code interprete

  • PAL: Program-aided Language Models. Preprint 2022 [pdf]

    Luyu Gao, Aman Madaan, Shuyan Zhou, Uri Alon, Pengfei Liu, Yiming Yang, Jamie Callan, Graham Neubig

    Annotation: code interprete

  • Solving Math Word Problems by Combining Language Models With Symbolic Solvers. Preprint 2022 [pdf]

    Joy He-Yueya, Gabriel Poesia, Rose E. Wang, Noah D. Goodman

    Annotation: code interprete

  • Faithful Chain-of-Thought Reasoning. Preprint 2023 [pdf]

    Qing Lyu, Shreya Havaldar, Adam Stein,Li Zhang, Delip Rao, Eric Wong, Marianna Apidianaki, Chris Callison-Burch

    Annotation: code interprete

  • Inner Monologue: Embodied Reasoning through Planning with Language Models. Preprint 2022 [pdf], [Blog]

    Wenlong Huang, Fei Xia, Ted Xiao, Harris Chan, Jacky Liang, Pete Florence, Andy Zeng, Jonathan Tompson, Igor Mordatch, Yevgen Chebotar, Pierre Sermanet, Noah Brown, Tomas Jackson, Linda Luu, Sergey Levine, Karol Hausman, Brian Ichter

    Annotation: virtual or physical

  • Chameleon: Plug-and-Play Compositional Reasoning with Large Language Models. Preprint 2023 [pdf], [Blog], [Code]

    Pan Lu, Baolin Peng, Hao Cheng, Michel Galley, Kai-Wei Chang, Ying Nian Wu, Song-Chun Zhu, Jianfeng Gao

    Annotation: utilize LLMs as a controller to coordinate multiple neural models in multimodal reasoning tasks

  • HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in HuggingFace. Preprint 2023 [pdf] [Code]

    Yongliang Shen, Kaitao Song, Xu Tan, Dongsheng Li, Weiming Lu, Yueting Zhuang

    Annotation: utilize LLMs as a controller to coordinate Hugging in multimodal reasoning tasks

  • ART: Automatic multi-step reasoning and tool-use for large language models. Preprint 2023 [pdf]

    Bhargavi Paranjape, Scott Lundberg, Sameer Singh, Hannaneh Hajishirzi, Luke Zettlemoyer, Marco Tulio Ribeiro

    Annotation: introduce ART, a framework that uses frozen LLMs to automatically generate reasoning steps as a program.

Clone this wiki locally