Junbo Zhao
教师
Works on large language models and representation learning, including white-box hallucination detection, demonstration validation for in-context learning, seed-free instruction tuning and robust cross-lingual named entity recognition.
Junbo Zhao's lab at Zhejiang works on large language models, interpretability, evaluation & safety of LMs and recognition, detection & segmentation. 64 papers since 2023, 47 at top venues, mostly Findings of ACL and EMNLP. Applications close .
At a glance
- papers, 2023–2026
- 17/ year
- at T1 venues
- 73%
- PhD students
- —
- MS students
- —
- alumni listed
- —
Papers by yearfrom DBLP, 2023–2026
- 13 papers2026 · T1 9 · collecting
- 21 papers2025 · T1 14
- 17 papers2024 · T1 12
- 13 papers2023 · T1 12
T1 venuesOtherVenues: Findings of ACL, EMNLP, ICLR, ACL, NeurIPS
Research focuspaper titles that use each topic's words
- Large language models26 papers
- Recognition, detection & segmentation4 papers
- Interpretability, evaluation & safety of LMs3 papers
- instruction tuning2 papers
- instruction2 papers
Paper list
From DBLP.
202613 papers
- AAAI (T1)An Invariant Latent Space Perspective on Language Model Inversion (opens in a new tab)
- Findings of ACLFine-Grained Data Ordering Improves Fine-Tuning for Large Language Models (opens in a new tab)
- Findings of ACLHSS-Synth: Humanities and Social Sciences Data Synthesis for LLMs (opens in a new tab)
- WWW (T1)KMLP: A Scalable Hybrid Architecture for Web-Scale Tabular Data Modeling (opens in a new tab)
- Machine LearningLADA: Label Disambiguation and Domain-Aware Learning for Domain Generalization (opens in a new tab)
- ICLR (T1)OptimSyn: Influence-Guided Rubrics Optimization for Synthetic Data Generation (opens in a new tab)
- TKDE (T1)Pretrained Model Recommendation for Downstream Fine-Tuning (opens in a new tab)
- WWW (T1)Reinforcement Learning with Verbalized Probabilities for LLM Classification (opens in a new tab)
- ICLR (T1)Stop Unnecessary Reflection: Training LRMs for Efficient Reasoning with Adaptive Reflection and Length Coordinated Penalty (opens in a new tab)
- WWW (T1)Toward real-world Table Agents: capabilities, workflows, and design principles for LLM-based table intelligence (opens in a new tab)
- ACL (T1)Towards Interpretable Tabular Reasoning: Enhancing LLM Reasoning on Tabular Data with Pre-Constructed Logic Graph (opens in a new tab)
- ICLR (T1)TraPO: A Semi-Supervised Reinforcement Learning Framework for Boosting LLM Reasoning (opens in a new tab)
202521 papers
- COLING (T1)AIGT: AI Generative Table Based on Prompt (opens in a new tab)
- Findings of ACLALPS: Attention Localization and Pruning Strategy for Efficient Adaptation of Large Language Models (opens in a new tab)
- ICLR (T1)Bridging the Semantic Gap Between Text and Table: A Case Study on NL2SQL (opens in a new tab)
- Machine LearningCaCOM: customizing text-to-image diffusion models in the wild via continual active selection (opens in a new tab)
- EMNLP (T1)CrowdAgent: Multi-Agent Managed Multi-Source Annotation System (opens in a new tab)
- EMNLP (T1)CYCLE-INSTRUCT: Fully Seed-Free Instruction Tuning via Dual Self-Training and Cycle Consistency (opens in a new tab)
- ACL (T1)D.Va: Validate Your Demonstration First Before You Use It (opens in a new tab)
- ICLR (T1)DataMan: Data Manager for Pre-training Large Language Models (opens in a new tab)
- EMNLP (T1)Ensembling Prompting Strategies for Zero-Shot Hierarchical Text Classification with Large Language Models (opens in a new tab)
- NeurIPS (T1)Harnessing Feature Resonance under Arbitrary Target Alignment for Out-of-Distribution Node Detection (opens in a new tab)
- Findings of NAACLJailbreaking Prompt Attack: A Controllable Adversarial Attack against Diffusion Models (opens in a new tab)
- ACL (T1)Large Margin Representation Learning for Robust Cross-lingual Named Entity Recognition (opens in a new tab)
202417 papers
- AAAI (T1)A Separation and Alignment Framework for Black-Box Domain Adaptation (opens in a new tab)
- VLDB (T1)CORAL: Collaborative Automatic Labeling System based on Large Language Models (opens in a new tab)
- Findings of ACLData Contamination Calibration for Black-box LLMs (opens in a new tab)
- Findings of ACLDORY: Deliberative Prompt Recovery for LLM (opens in a new tab)
- EMNLP (T1)Embedding and Gradient Say Wrong: A White-Box Method for Hallucination Detection (opens in a new tab)
- ICLR (T1)Energy-based Automated Model Evaluation (opens in a new tab)
- Findings of EMNLPFlowBench: Revisiting and Benchmarking Workflow-Guided Planning for LLM-based Agents (opens in a new tab)
- ACL (T1)Learning Geometry-Aware Representations for New Intent Discovery (opens in a new tab)
- NeurIPS (T1)Locating What You Need: Towards Adapting Diffusion Models to OOD Concepts In-the-Wild (opens in a new tab)
- Findings of ACLOn LLMs-Driven Synthetic Data Generation, Curation, and Evaluation: A Survey (opens in a new tab)
- TPAMI (T1)PiCO+: Contrastive Label Disambiguation for Robust Partial Label Learning (opens in a new tab)
- CVPR (T1)Positive-Unlabeled Learning by Latent Group-Aware Meta Disambiguation (opens in a new tab)
202313 papers
- ICCV (T1)CAME: Contrastive Automated Model Evaluation (opens in a new tab)
- WWW (T1)Catch: Collaborative Feature Set Search for Automated Feature Engineering (opens in a new tab)
- NeurIPS (T1)Debiased and Denoised Entity Recognition from Distant Supervision (opens in a new tab)
- EMNLP (T1)FreeAL: Towards Human-Free Active Learning in the Era of Large Language Models (opens in a new tab)
- Briefings Bioinform.Fuse feeds as one: cross-modal framework for general identification of AMPs (opens in a new tab)
- ICCV (T1)iDAG: Invariant DAG Searching for Domain Generalization (opens in a new tab)
- IJCAI (T1)Latent Processes Identification From Multi-View Time Series (opens in a new tab)
- ICLR (T1)Learning a Data-Driven Policy Network for Pre-Training Automated Feature Engineering (opens in a new tab)
- IJCAI (T1)ProMix: Combating Label Noise via Maximizing Clean Sample Utility (opens in a new tab)
- EMNLP (T1)Revisiting the Knowledge Injection Frameworks (opens in a new tab)
- VLDB (T1)SmartLite: A DBMS-based Serving System for DNN Inference in Resource-constrained Environments (opens in a new tab)
- NeurIPS (T1)SPA: A Graph Spectral Alignment Perspective for Domain Adaptation (opens in a new tab)
Where alumni went
Alumni are not listed on the lab homepage.
Apply to ZhejiangFall 2027
Computer Science PhD (international applicants)
- Round
- Self-funded
- Opens
- —
- GRE
- See program page
- English
- Per ZJU language requirements
- Fee
- RMB 800
Not yet confirmed for this intake.
Similar labsnearest by research summary
- SGNTU
Shafiq Joty
- Large language models
- CNSJTU
Yu Wang
- Large language models
- CNSJTU
Renqiu Xia
- Reasoning & agents
- CNSJTU
Mingda Chen
- Large language models
- USUW
Luke Zettlemoyer
- Large language models
- CNZhejiang
Tao Jin
- Multimodal & grounded language
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- The member list could not be found on the lab site.
- The alumni list could not be found on the lab site.
- The lab site and the paper record do not agree.
- Collected
- Verified
- Papers
- DBLP
- Members
- Not collected
- Alumni
- Not collected
Tiers are venue classifications, not lab rankings.
Zhejiangcloses