Xiangyu Peng (Becky)

Xiangyu Peng (Becky)

Senior Research Scientist @ Salesforce Research. Ph.D. in Machine Learning, Georgia Tech.

Publications

2026
MTA-Agent: An Open Recipe for Multimodal Deep Search Agents
Xiangyu Peng, Can Qin, An Yan, Xinyi Yang, Zeyuan Chen, Ran Xu, Chien-Sheng Wu
MLLMDeep SearchRL Training
MTA-Agent framework overview
arXiv GitHub
Scaling Knowledge Graph Construction through Synthetic Data Generation and Distillation
Prafulla Kumar Choubey, Xin Su, Man Luo, Xiangyu Peng, Caiming Xiong, Tiep Le, Shachar Rosenman, Vasudev Lal, Phil Mui, Ricky Ho, Phillip Howard, Chien-Sheng Wu
SynthKG framework overview
International Conference on Learning Representations (ICLR 2026)
arXiv
UniDoc-Bench: A Unified Benchmark for Document-Centric Multimodal RAG
Xiangyu Peng, Can Qin, Zeyuan Chen, Ran Xu, Caiming Xiong, Chien-Sheng Wu
MM-RAGBenchmarkingEvaluation
UniDoc-Bench overview
CVPR 2026 Workshop on Multimodal Agents and Retrieval (MAR)
arXiv GitHub HuggingFace
2025
Unanswerability Evaluation for Retrieval Augmented Generation
Xiangyu Peng, Prafulla Kumar Choubey, Caiming Xiong, Chien-Sheng Wu
RAGEvaluation
UAEval4RAG framework overview
Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)
ACL-25 GitHub
Turning Conversations into Workflows: A Framework to Extract and Evaluate Dialog Workflows for Service AI Agents
Prafulla Kumar Choubey, Xiangyu Peng, Shilpa Bhagavath, Caiming Xiong, Shiva Kumar Pentyala, Chien-Sheng Wu
Findings of the Association for Computational Linguistics: ACL 2025
ACL-25 Findings
Benchmarking Deep Search over Heterogeneous Enterprise Data
Prafulla Kumar Choubey, Xiangyu Peng, Shilpa Bhagavath, Kung-Hsiang Huang, Caiming Xiong, Chien-Sheng Wu
Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track (EMNLP 2025)
EMNLP-25 GitHub
ReGenesis: LLMs can Grow into Reasoning Generalists via Self-Improvement
Xiangyu Peng, Congying Xia, Xinyi Yang, Caiming Xiong, Chien-Sheng Wu, Chen Xing
ReasoningFine-tuning
ReGenesis framework overview
International Conference on Learning Representations (ICLR 2025) — Oral Presentation
arXiv GitHub ICLR-25 Oral
BingoGuard: LLM Content Moderation Tools with Risk Levels
Fan Yin, Philippe Laban, Xiangyu Peng, Yilun Zhou, Yixin Mao, Vaibhav Vats, Linnea Ross, Divyansh Agarwal, Caiming Xiong, Chien-Sheng Wu
International Conference on Learning Representations (ICLR 2025)
arXiv ICLR-25
Strefer: Empowering Video LLMs with Space-Time Referring and Reasoning via Synthetic Instruction Data
Honglu Zhou, Xiangyu Peng, Shrikant Kendre, Michael S. Ryoo, Silvio Savarese, Caiming Xiong, Juan Carlos Niebles
Video LLMSpace-Time-Aware ReasoningSynthetic Data Generation
Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV 2025)
ICCV-25
2023
Story Shaping: Teaching Agents Human-like Behavior with Stories
Xiangyu Peng, Christopher Cui, Wei Zhou, Renee Jia, Mark Riedl
Proceedings of the 19th AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE-23)
arXiv AIIDE-23
Neuro-Symbolic World Models for Adapting to Open World Novelty
Jonathan Balloch, Zhiyu Lin, Robert Wright, Xiangyu Peng, Mustafa Hussain, Aarun Srinivas, Julia Kim, Mark O. Riedl
Under Review
arXiv
Model Ensemble Instead Of Prompt Fusion: A Sample-Specific Knowledge Transfer Method For Few-Shot Prompt Tuning
Xiangyu Peng, Chen Xing, Prafulla Kumar Choubey, Chien-Sheng Wu, Caiming Xiong
International Conference on Learning Representations (ICLR-23)
arXiv ICLR-23
2022
XFBoost: Improving Text Generation with Controllable Decoders
Xiangyu Peng, Michael Sollami
U.S. Patent Application 17/509,024
arXiv
Inherently Explainable Reinforcement Learning in Natural Language
Xiangyu Peng, Mark O Riedl, Prithviraj Ammanabrolu
Thirty-sixth Conference on Neural Information Processing Systems (NeurIPS-22)
arXiv NeurIPS-22
Guiding Neural Story Generation with Reader Models
Xiangyu Peng, Kaige Xie, Amal Alabdulkarim, Harshith Kayam, Samihan Dani, Mark O. Riedl
Findings of the Association for Computational Linguistics: EMNLP 2022
arXiv EMNLP-22
Inferring the Reader: Guiding Automated Story Generation with Commonsense Reasoning
Xiangyu Peng, Siyan Li, Sarah Wiegreffe, Mark Riedl
Findings of the Association for Computational Linguistics: EMNLP 2022
arXiv EMNLP-22
NovGrid: A Flexible Grid World for Evaluating Agent Response to Novelty
Jonathan Balloch, Zhiyu Lin, Mustafa Hussain, Aarun Srinivas, Xiangyu Peng, Julia Kim, Mark Riedl
AAAI 2022 Spring Symposium on Designing Artificial Intelligence for Open Worlds
arXiv AAAI-22
2021
Automatic Story Generation: Challenges and Attempts
Amal Alabdulkarim*, Siyan Li*, Xiangyu Peng*
The 3rd Workshop On Narrative Understanding, NAACL-21
arXiv NAACL-21
Detecting and Adapting to Novelty in Games
Xiangyu Peng, Jonathan C. Balloch, Mark O. Riedl
Reinforcement Learning in Games workshop at AAAI 2021
arXiv AAAI-21
2020
Reducing Non-Normative Text Generation from Language Models
Xiangyu Peng, Siyan Li, Spencer Frazier, Mark Riedl
The 13th International Conference on Natural Language Generation (INLG 2020)
arXiv INLG-20