Tianyu Liu

Short Biography

Tianyu Liu is now a researcher at Alibaba’s Qwen Team (Beijing), working on the foundational Large Language Models (LLMs). He was a senior researcher at Tencent Cloud AI and contributed to the development of HuyuanAide (Tencent’s proprietary LLM series). Tianyu earned his PhD degree at Peking University in 2021 where he was advised by Zhifang Sui and Baobao Chang. During his PhD, he had wonderful experiences interning at or visiting Microsoft Research (Beijing and Redmond) and Toyota Technological Institute at Chicago (TTIC).

Before the waves of LLMs, his research experiences cover Natural Language Generation (e.g. data-to-text, summarization, free-form text, stories), Information Extraction (e.g. relation extraction, knowledge base population) and Robustness of NLP models. Begining from 2021, his research interests switch to the pretraining and alignment of large language models (e.g. code, reasoning, cross-modal, agent/tool-using, etc.).

News

• [Jan. 2024] We introduce Spec-Bench📖: A Comprehensive Benchmark and Unified Evaluation Platform for Speculative Decoding!🚀 Along with a comprehensive survey and review in Chinese. Check it out!
• [Oct. 2023] PCA-Eval is out! It’s an innovative benchmark for evaluating multi-domain embodied decision-making, specifically focusing on the performance in perception, cognition, and action. Check out our review in Chinese.
• [Oct. 2023] We have 4 papers accepted by EMNLP 2023 and its findings.
• [Jun. 2023] We have released an automatic and fairer evaluator FairEval for LLM comparisons that mitigate the systematic positional bias on the order of appearance in the context. Check out our paper or review in Chinese.
• [May. 2023] We have 2 papers accepted by ACL 2023 and its findings.
• [Oct. 2022] We have 4 papers accepted by EMNLP 2022 and its findings.
• [Apr. 2022] Our paper on Robust Finetuning is accepted by IJCAI 2022.
• [Apr. 2022] We have 3 papers accepted by NAACL 2022 and its findings.
• [Feb. 2022] We have 3 papers accepted by ACL 2022.

Publications [Google Scholar][DBLP]

* indicates equal contribution

Preprints

• Unlocking Efficiency in Large Language Model Inference: A Comprehensive Survey of Speculative Decoding
  Heming Xia, Zhe Yang, Qingxiu Dong, Peiyi Wang, Yongqi Li, Tao Ge, Tianyu Liu, Wenjie Li, Zhifang Sui
  [PDF] [bib] [Spec-Bench] [Paper List]

• Towards End-to-End Embodied Decision Making via Multi-modal Large Language Model: Explorations with GPT4-Vision and Beyond
  Liang Chen, Yichi Zhang, Shuhuai Ren, Haozhe Zhao, Zefan Cai, Yuchi Wang, Peiyi Wang, Tianyu Liu, Baobao Chang
  [PDF] [bib] [PCA-Eval]

• Making Large Language Models Better Reasoners with Alignment
  Peiyi Wang, Lei Li, Liang Chen, Feifan Song, Binghuai Lin, Yunbo Cao, Tianyu Liu, Zhifang Sui
  [PDF] [bib]

• Large Language Models are not Fair Evaluators
  Peiyi Wang, Lei Li, Liang Chen, Dawei Zhu, Binghuai Lin, Yunbo Cao, Qi Liu, Tianyu Liu, Zhifang Sui
  [PDF] [bib] [FairEval]

• RepCL: Exploring Effective Representation for Continual Text Classification
  Yifan Song*, Peiyi Wang*, Dawei Zhu, Tianyu Liu, Zhifang Sui, Sujian Li
  [PDF] [bib]

• DialogVCS: Robust Natural Language Understanding in Dialogue System Upgrade
  Zefan Cai*, Xin Zheng*, Tianyu Liu*, Xu Wang, Haoran Meng, Jiaqi Han, Gang Yuan, Binghuai Lin, Baobao Chang, Yunbo Cao
  [PDF] [bib]

• Less is More: Rethinking State-of-the-art Continual Relation Extraction Models with a Frustratingly Easy but Effective Approach
  Peiyi Wang*, Yifan Song*, Tianyu Liu, Rundong Gao, Binghuai Lin, Yunbo Cao, Zhifang Sui
  [PDF] [bib]

• SmartSales: Sales Script Extraction and Analysis from Sales Chatlog
  Hua Liang*, Tianyu Liu*, Peiyi Wang, Mengliang Rao, Yunbo Cao
  [PDF] [bib]

• Explicit Interaction Network for Aspect Sentiment Triplet Extraction
  Peiyi Wang, Tianyu Liu, Damai Dai, Runxin Xu, Baobao Chang, Zhifang Sui
  [PDF] [bib]

• First target and opinion then polarity: Enhancing target-opinion correlation for aspect sentiment triplet extraction
  Lianzhe Huang*, Peiyi Wang*, Sujian Li, Tianyu Liu, Xiaodong Zhang, Zhicong Cheng, Dawei Yin, Houfeng Wang
  [PDF] [bib]

Peer-reviewed Papers

• ImageNetVC: Zero- and Few-Shot Visual Commonsense Evaluation on 1000 ImageNet Categories
  Heming Xia, Qingxiu Dong, Lei Li, Jingjing Xu, Tianyu Liu, Ziwei Qin, Zhifang Sui
  Findings of Empirical Methods in Natural Language Processing (Findings of EMNLP), 2023
  [PDF] [bib] [code]

• Bi-Drop: Enhancing Fine-tuning Generalization via Synchronous sub-net Estimation and Optimization
  Shoujie Tong, Heming Xia, Damai Dai, Runxin Xu, Tianyu Liu, Binghuai Lin, Yunbo Cao, Zhifang Sui
  Findings of Empirical Methods in Natural Language Processing (Findings of EMNLP), 2023
  [PDF] [bib] [code]

• DialogQAE: N-to-N Question Answer Pair Extraction from Customer Service Chatlog
  Xin Zheng*, Tianyu Liu*, Haoran Meng*, Xu Wang, Yufan Jiang, Mengliang Rao, Binghuai Lin, Zhifang Sui, Yunbo Cao
  Findings of Empirical Methods in Natural Language Processing (Findings of EMNLP), 2023
  [PDF] [bib] [code]

• InfoCL: Alleviating Catastrophic Forgetting in Continual Text Classification from An Information Theoretic Perspective
  Yifan Song, Peiyi Wang, Weimin Xiong, Dawei Zhu, Tianyu Liu, Zhifang Sui, Sujian Li
  Findings of Empirical Methods in Natural Language Processing (Findings of EMNLP), 2023
  [PDF] [bib] [code]

• Denoising Bottleneck with Mutual Information Maximization for Video Multimodal Fusion
  Shaoxiang Wu, Damai Dai, Ziwei Qin, Tianyu Liu, Binghuai Lin, Yunbo Cao, Zhifang Sui
  Association for Computational Linguistics (ACL), 2023
  [PDF] [bib] [code]

• Enhancing Continual Relation Extraction via Classifier Decomposition
  Heming Xia, Peiyi Wang, Tianyu Liu, Binghuai Lin, Yunbo Cao, Zhifang Sui
  Findings of Association for Computational Linguistics (ACL), 2023
  [PDF] [bib] [code]

• HPT: Hierarchy-aware Prompt Tuning for Hierarchical Text Classification
  Zihan Wang*, Peiyi Wang*, Tianyu Liu, Binghuai Lin, Yunbo Cao, Zhifang Sui, Houfeng Wang
  Empirical Methods in Natural Language Processing (EMNLP), 2022
  [PDF] [bib] [code]

• Learning Robust Representations for Continual Relation Extraction via Adversarial Class Augmentation
  Peiyi Wang*, Yifan Song*, Tianyu Liu, Binghuai Lin, Yunbo Cao, Sujian Li, Zhifang Sui
  Empirical Methods in Natural Language Processing (EMNLP), 2022
  [PDF] [bib] [code]

• DialogUSR: Complex Dialogue Utterance Splitting and Reformulation for Multiple Intent Detection
  Haoran Meng*, Zheng Xin*, Tianyu Liu*, Zizhen Wang, He Feng, Binghuai Lin, Xuemin Zhao, Yunbo Cao, Zhifang Sui
  Findings of Empirical Methods in Natural Language Processing (Findings of EMNLP), 2022
  [PDF] [bib] [code]

• Learning Invariant Representation Improves Robustness for MRC Models
  Hai Yu, Liang Wen, Haoran Meng, Tianyu Liu, Houfeng Wang
  Findings of Empirical Methods in Natural Language Processing (Findings of EMNLP), 2022
  [PDF] [bib]

• Robust Fine-tuning via Perturbation and Interpolation from In-batch Instances
  Shoujie Tong*, Qingxiu Dong*, Damai Dai, Tianyu Liu, Baobao Chang, Zhifang Sui
  International Joint Conference on Artificial Intelligence (IJCAI), 2022
  [PDF] [bib] [code]

• Hierarchical Curriculum Learning for AMR Parsing
  Peiyi Wang*, Liang Chen*, Tianyu Liu, Damai Dai, Yunbo Cao, Baobao Chang, Zhifang Sui
  Association for Computational Linguistics (ACL), 2022
  [PDF] [bib] [code]

• Premise-based Multimodal Reasoning: Conditional Inference on Joint Textual and Visual Clues
  Qingxiu Dong*, Ziwei Qin*, Heming Xia, Tian Feng, Shoujie Tong, Haoran Meng, Lin Xu, Zhongyu Wei, Weidong Zhan, Baobao Chang, Sujian Li, Tianyu Liu, Zhifang Sui
  Association for Computational Linguistics (ACL), 2022
  [PDF] [bib] [Leaderboard]

• A Token-level Reference-free Hallucination Detection Benchmark for Free-form Text Generation
  Tianyu Liu, Yizhe Zhang, Chris Brockett, Yi Mao, Zhifang Sui, Weizhu Chen, Bill Dolan
  Association for Computational Linguistics (ACL), 2022
  [PDF] [bib] [code]

• An Enhanced Span-based Decomposition Method for Few-Shot Sequence Labeling
  Peiyi Wang*, Runxin Xu*, Tianyu Liu, Qingyu Zhou, Yunbo Cao, Baobao Chang, Zhifang Sui
  North American Association for Computational Linguistics (NAACL), 2022
  [PDF] [bib] [code]

• A Two-Stream AMR-enhanced Model for Document-level Event Argument Extraction
  Runxin Xu*, Peiyi Wang*, Tianyu Liu, Shuang Zeng, Baobao Chang, Zhifang Sui
  North American Association for Computational Linguistics (NAACL), 2022
  [PDF] [bib] [code]

• ATP: AMRize Then Parse! Enhancing AMR Parsing with PseudoAMRs
  Liang Chen*, Peiyi Wang*, Runxin Xu, Tianyu Liu, Zhifang Sui, Baobao Chang
  Findings of North American Association for Computational Linguistics (Findings of NAACL), 2022
  [PDF] [bib] [code]

• Leveraging Word-Formation Knowledge for Chinese Word Sense Disambiguation
  Hua Zheng*, Lei Li*, Damai Dai, Deli Chen, Tianyu Liu, Xu Sun, Yang Liu
  Findings of Empirical Methods in Natural Language Processing (Findings of EMNLP), 2021
  [PDF] [bib] [code]

• Inductively Representing Out-of-Knowledge-Graph Entities by Optimal Estimation Under Translational Assumptions
  Damai Dai, Hua Zheng, Fuli Luo, Pengcheng Yang, Tianyu Liu, Zhifang Sui, Baobao Chang
  The 6th Workshop on Representation Learning for NLP (RepL4NLP), 2021
  [PDF] [bib] [code]

• Document-level Event Extraction via Heterogeneous Graph-based Interaction Model with a Tracker
  Runxin Xu, Tianyu Liu, Lei Li, Baobao Chang
  Association for Computational Linguistics (ACL), 2021
  [PDF] [bib] [code]

• Decompose, Fuse and Generate: A Formation-Informed Method for Chinese Definition Generation
  Hua Zheng*, Damai Dai*, Lei Li, Tianyu Liu, Zhifang Sui, Baobao Chang, Yang Liu
  North American Association for Computational Linguistics (NAACL), 2021
  [PDF] [bib] [code]

• Behind the Scenes: An Exploration of Trigger Biases Problem in Few-Shot Event Classification
  Peiyi Wang*, Runxin Xun*, Tianyu Liu, Damai Dai, Baobao Chang, Zhifang Sui
  ACM International Conference on Information & Knowledge Management (CIKM), 2021
  [PDF] [bib] [code]

• Towards Faithfulness in Open Domain Table-to-text Generation from an Entity-centric View
  Tianyu Liu*, Xin Zheng*, Baobao Chang, Zhifang Sui
  AAAI Conference on Artificial Intelligence (AAAI), 2021
  [PDF] [bib]

• An Anchor-Based Automatic Evaluation Metric for Document Summarization
  Kexiang Wang*, Tianyu Liu*, Baobao Chang, Zhifang Sui
  International Conference on Computational Linguistics (COLING), 2020
  [PDF] [bib]

• An Empirical Study on Model-agnostic Debiasing Strategies for Robust Natural Language Inference
  Tianyu Liu*, Xin Zheng*, Xiaoan Ding, Baobao Chang, Zhifang Sui
  Conference on Computational Natural Language Learning (CoNLL), 2020
  [PDF] [bib] [Adversarial Datasets]

• Discrimatively-Tuned Generative Classifier for Robust Natural Language Inference
  Xiaoan Ding*, Tianyu Liu*, Baobao Chang, Zhifang Sui, Kevin Gimpel
  Empirical Methods in Natural Language Processing (EMNLP), 2020
  [PDF] [bib]

• An Exploration of Arbitrary-Order Sequence Labeling via Energy-Based Inference Networks
  Lifu Tu*, Tianyu Liu*, Kevin Gimpel
  Empirical Methods in Natural Language Processing (EMNLP), 2020
  [PDF] [bib] [code]

• HypoNLI: Exploring the Artificial Patterns of Hypothesis-only Bias in Natural Language Inference
  Tianyu Liu, Xin Zheng, Baobao Chang, Zhifang Sui
  International Conference on Language Resources and Evaluation (LREC), 2020
  [PDF] [dataset(google drive)] [dataset(tencent cloud)]

• Towards Comprehensive Description Generation from Factual Attribute-Value Tables
  Tianyu Liu, Fuli Luo, Pengcheng Yang, Wei Wu, Baobao Chang, Zhifang Sui
  Association for Computational Linguistics (ACL), 2019
  [PDF] [bib]

• Key Fact as Pivot: A Two-Stage Model for Low Resource Table-to-Text Generation
  Shuming Ma, Pengcheng Yang, Tianyu Liu, Peng Li, Jie Zhou, Xu Sun
  Association for Computational Linguistics (ACL), 2019
  [PDF] [bib]

• Enhancing Topic-to-Essay Generation with External Commonsense Knowledge
  Pengcheng Yang*, Lei Li*, Fuli Luo, Tianyu Liu, Xu Sun
  Association for Computational Linguistics (ACL), 2019
  [PDF] [bib] [code]

• Learning to Control the Fine-grained Sentiment for Story Ending Generation
  Fuli Luo*, Damai Dai*, Pengcheng Yang, Tianyu Liu, Baobao Chang, Zhifang Sui, Xu Sun
  Association for Computational Linguistics (ACL), 2019
  [PDF] [bib] [code]

• MAAM: A Morphology-Aware Alignment Model for Unsupervised Bilingual Lexicon Induction
  Pengcheng Yang, Fuli Luo, Peng Chen, Tianyu Liu, Xu Sun
  Association for Computational Linguistics (ACL), 2019
  [PDF] [bib]

• Hierarchical Encoder with Auxiliary Supervision for Neural Table-to-text Generation: Learning Better Representation for Tables
  Tianyu Liu, Fuli Luo, Qiaolin Xia, Shuming Ma, Baobao Chang, Zhifang Sui
  AAAI Conference on Artificial Intelligence (AAAI), 2019
  [PDF] [bib]

• Leveraging Gloss Knowledge in Neural Word Sense Disambiguation by Hierarchical Co-Attention
  Fuli Luo, Tianyu Liu, Zexue He, Qiaolin Xia, Baobao Chang, Zhifang Sui
  Empirical Methods in Natural Language Processing (EMNLP), 2018
  [PDF] [bib]

• Phrase-level Self-Attention Networks for Universal Sentence Encoding
  Wei Wu, Houfeng Wang, Tianyu Liu, Shuming Ma
  Empirical Methods in Natural Language Processing (EMNLP), 2018
  [PDF] [bib]

• Incorporating Glosses into Neural Word Sense Disambiguation
  Fuli Luo, Tianyu Liu, Qiaolin Xia, Baobao Chang, Zhifang Sui
  Association for Computational Linguistics (ACL), 2018
  [PDF] [bib] [code]

• Table-to-text Generation by Structure-aware Seq2seq Learning
  Tianyu Liu, Kexiang Wang, Lei Sha, Baobao Chang, Zhifang Sui
  AAAI Conference on Artificial Intelligence (AAAI), 2018
  [PDF] [bib] [code]

• Order-Planning Neural Text Generation From Structured Data
  Lei Sha, Lili Mou, Tianyu Liu, Pascal Poupart, Sujian Li, Baobao Chang, Zhifang Sui
  AAAI Conference on Artificial Intelligence (AAAI), 2018
  [PDF] [bib]

• A Soft-label Method for Noise-tolerant Distantly Supervised Relation Extraction
  Tianyu Liu, Kexiang Wang, Baobao Chang, Zhifang Sui
  Empirical Methods in Natural Language Processing (EMNLP), 2017
  [PDF] [bib] [code]

• Affinity-Preserving Random Walk for Multi-Document Summarization
  Kexiang Wang, Tianyu Liu, Zhifang Sui, Baobao Chang
  Empirical Methods in Natural Language Processing (EMNLP), 2017
  [PDF] [bib]

• Large-scale Simple Question Generation by Template-based Seq2seq Learning
  Tianyu Liu, Bingzhen Wei, Baobao Chang, Zhifang Sui
  CCF International Conference on Natural Language Processing and Chinese Computing (NLPCC), 2017
  [paper] [bib] [code] Best Student Paper

• Encoding Temporal Information for Time-Aware Link Prediction
  Tingsong Jiang, Tianyu Liu, Tao Ge, Lei Sha, Sujian Li, Baobao Chang, Zhifang Sui
  Empirical Methods in Natural Language Processing (EMNLP), 2016
  [PDF] [bib]

• Towards Time-Aware Knowledge Graph Completion
  Tingsong Jiang, Tianyu Liu, Tao Ge, Lei Sha, Baobao Chang, Sujian Li, Zhifang Sui
  International Conference on Computational Linguistics (COLING), 2016
  [paper] [bib]

Experiences

• Sep 2023 - Now, Researcher, Qwen Team, Alibaba.
• Jul 2021 - Sep 2023, Senior Researcher, Cloud AI, Tencent.
• Jun 2020 - Nov 2020, Research Intern, Microsoft Research AI (Redmond).
  Mentors: Yizhe Zhang, Yi Mao
• Jul 2019 - Feb 2020, Visitor, Toyota Technological Institute at Chicago (TTIC)
  Mentors: Kevin Gimpel, Sam Wiseman
• Apr 2018 - Oct 2018, Research Intern, Knowledge Computing Group, Microsoft Research Asia.
  Mentors: Jin-Ge Yao, Chin-Yew Lin
• Sep 2016 - Jul 2021, PhD, Institute of Computational Linguistics (ICL), Peking University.
  Supervisors: Zhifang Sui, Baobao Chang
  Highlights: 2021 CIPS Outstanding Doctoral Dissertation Award (Runner up, top 8 nationwide)
• Sep. 2012 - Jul 2016, BE, School of Computer Science, Sichuan University.
  Mentor: Jiancheng Lv
  Highlights: 1st in GPA ranking ; Wu Yuzhang Honors Degree

Honors and Awards

• CIPS Outstanding Doctoral Dissertation Award (Runner up), 2021
• EMNLP Oustanding Reviewer, 2020
• Specialized Scholarship for Ph.D. Candidates, 2019
• Leo KoGuan Scholarship, 2018
• President Scolarship of PKU, 2018-2020
• Best Student Paper Award of NLPCC, 2017
• Acadamic Excellence Award of PKU, 2017-2020
• Outstanding Graduate Award of SCU, 2016
• Outstanding Thesis Award of SCU, 2016
• Honorable Mention in Mathematical Contest in Modeling, 2014
• Tang Lixing Scolarship, 2014-2019
• National Scolarship, 2013, 2017

Services

• Conference Program Committee Member:
  • Association for Computational Linguistics (ACL)
  • Empirical Methods in Natural Language Processing (EMNLP) [EMNLP-22 Outstanding Reviewer]
  • North American Association for Computational Linguistics (NAACL)
  • European Chapter of the Association for Computational Linguistics (EACL)
  • International Conference on Computational Linguistics (COLING)
  • International Conference on Natural Language Generation (INLG)
  • China National Conference on Computational Linguistics (CCL)
  • International Conference on Learning Representations (ICLR)
  • Conference on Language Modeling (COLM)
  • Association for the Advancement of Artificial Intelligence (AAAI)
  • International Joint Conference on Artificial Intelligence (IJCAI) [IJCAI 2021 Senior PC Member]
• Journal Program Committee Member:
  • IEEE Transactions on Audio, Speech, and Language Processing (TASLP)
  • IEEE Transactions on Knowledge and Data Engineering (TKDE)
• Teaching Assistant: Computational Linguistics (Lecturer: Baobao Chang), Compiler Principles (Lecturer: Zhifang Sui)