Jianguo Zhang 张建国 Ph.D in Computer Science University of Illinois at Chicago (UIC), Chicago, IL, U.S. Email: jianguozhang@salesforce.com (Note: jzhan51@uic.edu is deactivated by UIC)

I am a Senior Research Scientist at Salesforce AI Research and a recipient of the 2024 Annual T&P (Technology and Product) All-Star Awards, awarded to the top 1% with approximately 150 winners across all Technology. I earned my Ph.D. in Computer Science from the University of Illinois at Chicago, where I was advised by Prof. Philip S. Yu in the BDSC Lab. My current research primarily focuses on AI Agents and Conversational AI. Specifically, my work involves:

Large-scale and advanced data collection, processing, synthesis, verification, and more.

Training large action models and designing AI agents.

Creating end-to-end dialogue systems, with an emphasis on accurate natural language understanding and dialogue state tracking without complicated pipelines.

News

Apr 2025: 🏆 Check out xLAM-2-fc-r, which achieved Top-1 performance on the latest BFCL Leaderboard!

Apr 2025: 🚀 ActionStudio is now fully open-source! Dive into the details in our paper and explore the codebase for end-to-end agent orchestration.

Apr 2025: 📢 Discover APIGen-MT, our newly open-sourced framework for multi-turn data generation. Learn more on the project website!

Sep 2024: 🏆 Check our latest work xLAM: A Family of Large Action Models to Empower AI Agent Systems, which demonstrates top-tier performance across multiple challenging benchmarks and scenarios. Explore more through the xLAM GitHub Repo, HuggingFace Model Collections and Project Webpage.

July 2024: Happy to contribute to the work Consent in Crisis: The Rapid Decline of the AI Data Commons. See also the New York Times and Twitter.

June 2024: Check our new work APIGen, the best open-sourced models for function calling. Our dataset xlam-function-calling-60k is currently among the Top-2 trending datasets on HuggingFace, standing out in a field of 176,238 datasets as of July 11, 2024. See also the Twitter by Salesforce CEO, VentureBeat and 新智元.

March 2024: xLAM Repository and xLAM v0.1 Model are released! Try it together with AgentLite benchmark or other benchamrks, which is comparable to GPT-4! See also the Paper and Twitter by Salesforce AI Research VP.

Dec 2023: DialogStudio Repository has garnered 400 stars. DialogStudio Data has more than 55k downloads and was previously ranked among the Top-10 trending datasets! See also the Paper and Twitter.

Jan 2022: I will join Salesforce AI Research as a Research Scientist.

Sep 2021: Finished my internship at Facebook AI Research (FAIR). The project focuses on few-shot end-to-end task-oriented dialgue systems.

Aug 2021: Two paper are accepted by EMNLP 2021 main conference as short papers. One first-author work collaborates with Adobe Research and it focuses on natural language understanding.

June 2021: Check our new released paper and public resources for few-shot intent detection w/ and w/o out-of-scope queries.

May 2021: I finished my internship at Adobe Research and submitted one research paper, where I also discover PhotoShop user intents and improve intent detection performance using large-scale unlabeled PhotoShop user queries.

Dec 2020: Our toolkit NaturalCC has been released in GitHub, which can be accessed via the Homepage.

Nov 2020: We have released the paper, video for the EMNLP 2020 paper, and the corresponding code is available here!

Sep 2020: One first-author work collaborated with Salesforce Research is accepted by *SEM 2020 as a long paper. The work focuses on multi-domain dialog state tracking and it originally got borderline scores on ACL 2020. We have updated the paper!

Sep 2020: One first-author work collaborated with Salesforce Research is accepted by EMNLP 2020 main conference as a long paper. The work focuses on natural language understanding.

June 2020: One work collaborated with Google Research is accepted by ACL 2020 NLP4ConvAI, and the modified high-quality version of MultiWOZ 2.2 dataset with additional annotation corrections and state tracking baselines is released here!

Open Sources

xLAM | AgentLite | DialogStudio | MultiWOZ |Data Provenance | NaturalCC

Publications

Last Update: Aug. 2024

• Bridging the Data Provenance Gap Across Text, Speech and Video  

        Shayne Longpre et al. Jianguo Zhang (Top Contributor)

        🏆 ICLR 2025 & Featured by Top Media

        [Paper] / [Project Website] / [GitHub]

• xLAM: A Family of Large Action Models to Empower AI Agent Systems  

        Jianguo Zhang, Tian Lan, Ming Zhu, Zuxin Liu, Thai Hoang, Shirley Kokane, Weiran Yao, Juntao Tan, Akshara Prabhakar, Haolin Chen, Zhiwei Liu, Yihao Feng, Tulika Awalgaonkar, Rithesh Murthy, Eric Hu, Zeyuan Chen, Ran Xu, Juan Carlos Niebles, Shelby Heinecke, Huan Wang, Silvio Savarese, Caiming Xiong

        🏆 NAACL 2025 Oral & Featured by Salesforce CEO & Top Media

        [Paper] /[GitHub] /[Models]

• TACO: Learning Multi-modal Action Models with Synthetic Chains-of-Thought-and-Action  

        Zixian Ma, Jianguo Zhang, Zhiwei Liu, Jieyu Zhang, Juntao Tan, Manli Shu, Juan Carlos Niebles, Shelby Heinecke, Huan Wang, Caiming Xiong, Ranjay Krishna, Silvio Savarese

        [ArXiv Paper] /[Project Website] /[Data] /[Code]

• PRACT: Optimizing Principled Reasoning and Acting of LLM Agent  

        Zhiwei Liu, Weiran Yao, Jianguo Zhang, Rithesh Murthy, Liangwei Yang, Zuxin Liu, Tian Lan, Ming Zhu, Juntao Tan, Shirley Kokane, Thai Hoang, Juan Carlos Niebles, Shelby Heinecke, Huan Wang, Silvio Savarese, Caiming Xiong

        CoNLL 2024

        [Paper]

• Apigen: Automated pipeline for generating verifiable and diverse function-calling datasets  

        Zuxin Liu, Thai Hoang, Jianguo Zhang, Ming Zhu, Tian Lan, Shirley Kokane, Juntao Tan, Weiran Yao, Zhiwei Liu, Yihao Feng, Rithesh Murthy, Liangwei Yang, Silvio Savarese, Juan Carlos Niebles, Huan Wang, Shelby Heinecke, Caiming Xiong

        🏆 NeurIPS 2024 Datasets and Benchmarks Track & Featured by Top Media

        [Paper] / [Project Website] / [GitHub] /[Data]

• Consent in Crisis: The Rapid Decline of the AI Data Commons  

        Shayne Longpre et al. Jianguo Zhang (Contributor)

        🏆 NeurIPS 2024 Datasets and Benchmarks Track & Featured by Top Media

        [Paper] / [Project Website] / [GitHub]

• MobileAIBench: Benchmarking LLMs and LMMs for On-Device Use Cases  

        Rithesh Murthy, Liangwei Yang, Juntao Tan, Tulika Manoj Awalgaonkar, Yilun Zhou, Shelby Heinecke, Sachin Desai, Jason Wu, Ran Xu, Sarah Tan, Jianguo Zhang, Zhiwei Liu, Shirley Kokane, Zuxin Liu, Ming Zhu, Huan Wang, Caiming Xiong, Silvio Savarese

        [ArXiv Paper] / [GitHub]

• AgentLite: A Lightweight Library for Building and Advancing Task-Oriented LLM Agent System  

        Zhiwei Liu, Weiran Yao, Jianguo Zhang, Liangwei Yang, Zuxin Liu, Juntao Tan, Prafulla K Choubey, Tian Lan, Jason Wu, Huan Wang, Shelby Heinecke, Caiming Xiong, Silvio Savarese

        🏆 Databricks Data + AI Summit

        [Paper] / [GitHub]

• AgentOhana: Design Unified Data and Training Pipeline for Effective Agent Learning  

        Jianguo Zhang, Tian Lan, Rithesh Murthy, Zhiwei Liu, Weiran Yao, Juntao Tan, Thai Hoang, Liangwei Yang, Yihao Feng, Zuxin Liu, Tulika Awalgaonkar, Juan Carlos Niebles, Silvio Savarese, Shelby Heinecke, Huan Wang, Caiming Xiong

        🏆 Hugging Face Daily Papers

        [Paper] / [GitHub] / [Models]

• DRDT: Dynamic Reflection with Divergent Thinking for LLM-based Sequential Recommendation  

        Yu Wang, Zhiwei Liu, Jianguo Zhang, Weiran Yao, Shelby Heinecke, Philip S. Yu

        [ArXiv Paper]

• Deep Learning for Code Intelligence: Survey, Benchmark and Toolkit  

        Yao Wan, Yang He, Zhangqian Bi, Jianguo Zhang, Hongyu Zhang, Yulei Sui, Guandong Xu, Hai Jin, Philip S. Yu.

        ACM Computing Surveys, 2024

        [Paper] / [Project Website]

• BOLAA: Benchmarking and orchestrating LLM-augmented autonomous agents  

        Zhiwei Liu, Weiran Yao, Jianguo Zhang, Le Xue, Shelby Heinecke, Rithesh Murthy, Yihao Feng, Zeyuan Chen, Juan Carlos Niebles, Devansh Arpit, Ran Xu, Phil Mui, Huan Wang, Caiming Xiong, Silvio Savarese

        🏆 Hugging Face Daily Papers

        [Paper] / [GitHub]

• DialogStudio: Towards Richest and Most Diverse Unified Dataset Collection for Conversational AI  

        Jianguo Zhang*, Kun Qian*, Zhiwei Liu, Shelby Heinecke, Rui Meng, Ye Liu, Zhou Yu, Silvio Savarese, Caiming Xiong

        🏆 Top 10 Trending Hugging Face Datasets in Sep. 2023 & Findings of EACL 2024

        [Paper] / [GitHub] / [HuggingFace]

• Retroformer: Retrospective large language agents with policy gradient optimization  

        Weiran Yao, Shelby Heinecke, Juan Carlos Niebles, Zhiwei Liu, Yihao Feng, Le Xue, Rithesh Murthy, Zeyuan Chen, Jianguo Zhang, Devansh Arpit, Ran Xu, Phil Mui, Huan Wang, Caiming Xiong, Silvio Savarese

        🏆 ICLR 2024 - Spotlight + Hugging Face #1 Paper of the Day

        [Paper] / [GitHub]

• Enhancing Performance on Seen and Unseen Dialogue Scenarios using Retrieval-Augmented End-to-End Task-Oriented System  

        Jianguo Zhang, Stephen Roller, Kun Qian, Zhiwei Liu, Rui Meng, Shelby Heinecke, Huan Wang, Silvio Savarese, Caiming Xiong

        SIGDIAL 2023

        [Paper]

• Fantastic Rewards and How to Tame Them: A Case Study on Reward Learning for Task-Oriented Dialogue Systems  

        Yihao Feng, Shentao Yang, Shujian Zhang, Jianguo Zhang, Caiming Xiong, Mingyuan Zhou, Huan Wang

        ICLR 2023

        [Paper]

• Zero-shot Item-based Recommendation via Multi-task Product Knowledge Graph Pre-Training  

        Ziwei Fan, Zhiwei Liu, Shelby Heinecke, Jianguo Zhang, Huan Wang, Caiming Xiong, Philip S Yu

        CIKM 2023

        [Paper]

• Are Pretrained Transformers Robust in Intent Classification? A Missing Ingredient in Evaluation of Out-of-Scope Intent Detection  

        Jianguo Zhang, Kazuma Hashimoto, Yao Wan, Zhiwei Liu, Ye Liu, Caiming Xiong, Philip Yu

        ACL 2022 Workshop on NLP for Conversational AI

        [Paper] [GitHub]

• NaturalCC: An Open-Source Toolkit for Code Intelligence  

        Yao Wan, Yang He, Zhangqian Bi, Jianguo Zhang, Yulei Sui, Hongyu Zhang, Kazuma Hashimoto, Hai Jin, Guandong Xu, Caiming Xiong and Philip Yu

        ICSE 2022 Demo Track

        [Toolkit]

• Few-Shot Intent Detection via Contrastive Pre-Training and Fine-Tuning  

        Jianguo Zhang, Trung Bui, Seunghyun Yoon, Xiang Chen, Zhiwei Liu, Congying Xia, Quan Hung Tran, Walter Chang and Philip Yu

        EMNLP 2021 (Oral)

        [Paper]

• HETFORMER: Heterogeneous Transformer with Sparse Attention for Long-Text Extractive Summarization

        Ye Liu, Jianguo Zhang, Yao Wan, Congying Xia, Lifang He and Philip Yu

        EMNLP 2021(Oral)

        [Paper]

• Enriching Non-Autoregressive Transformer with Syntactic and Semantic Structures for Neural Machine Translation  

        Ye Liu, Yao Wan, Jianguo Zhang, Wenting Zhao, Philip Yu

        EACL 2021

        [Paper]

• Discriminative Nearest Neighbor Few-Shot Intent Detection by Transferring Natural Language Inference  

        Jianguo Zhang, Kazuma Hashimoto, Wenhao Liu, Chien-Sheng Wu, Yao Wan, Philip S Yu, Richard Socher, Caiming Xiong

        EMNLP 2020

        [Paper] / [Slide] / [Video] / [Code]

• Find or Classify? Dual Strategy for Slot-Value Predictions on Multi-Domain Dialog State Tracking  

        Jianguo Zhang, Kazuma Hashimoto, Chien-Sheng Wu, Yao Wan, Philip S Yu, Richard Socher, Caiming Xiong

        *SEM 2020(Oral)

        [Paper] / [Slide] / [Video]

• MultiWOZ 2.2 : A Dialogue Dataset with Additional Annotation Corrections and State Tracking Baselines  

        Xiaoxue Zang, Abhinav Rastogi, Srinivas Sunkara, Raghav Gupta, Jianguo Zhang, Jindong Chen

        ACL 2020 Workshop on NLP for Conversational AI

        [Paper] / [Dataset ][Video]

• Multi-Modal Generative Adversarial Network for Short Product Title Generation in Mobile E-Commerce  

        Jianguo Zhang, Pengcheng Zou, Zhao Li, Yao Wan, Xiuming Pan, Yu Gong, Philip S Yu

        NAACL-HLT 2019 (Oral)

        [Paper]

• Product Title Refinement via Multi-Modal Generative Adversarial Learning  

        Jianguo Zhang, Pengcheng Zou, Zhao Li, Yao Wan, Ye Liu, Xiuming Pan, Yu Gong, Philip S Yu

        NeurIPS 2018 Workshop on ViGIL

        [Paper]

• Layerwise Perturbation-Based Adversarial Training for Hard Drive Health Degree Prediction  

        Jianguo Zhang*, Ji Wang*, Lifang He, Zhao Li, Philip S Yu (* indicates equal contribution)

        ICDM 2018

        [Paper]

• Not just privacy: Improving performance of private deep learning in mobile cloud  

        Ji Wang*, Jianguo Zhang* , Weidong Bao, Xiaomin Zhu, Bokai Cao, Philip S Yu (* indicates equal contribution)

        KDD 2018

        [Paper]

Service

Mentoring Interns at Salesforce: Zixian Ma (2024), Liangwei Yang (2023), Ziwei Fan (2022)

Program Committee/Reviewer: ACL Rolling Review 2021-pres, EMNLP 2021-Pres, NeurIPS 2023-Pres, EACL 2023, ACL 2021/2022, NAACL 2021, IEEE/ACM TASLP 2020-Pres, and others.

Session Chair: COLING 2020 (industry track: Dialogue).

Senior Program Committee: CIKM 2023-Pres.

This website is under construction