Jinghui Lu
Scholar

Jinghui Lu

Google Scholar ID: ZzK_UdYAAAAJ
ByteDance Inc., School of Computer Science, University College Dublin
Natural Language ProcessingMulti-ModalityLLMHuman-in-the-loop Learning
Citations & Impact
All-time
Citations
588
 
H-index
13
 
i10-index
20
 
Publications
20
 
Co-authors
16
list available
Resume (English only)
Academic Achievements
  • Published 20+ papers at top-tier AI/NLP conferences including NeurIPS, ACL, NAACL, AAAI, and COLING
  • 2025: 4 papers accepted to ACL 2025; 2 to EMNLP 2025; 1 to ICCV 2025
  • 2024: First-author paper at NeurIPS 2024; papers at ACL 2024, NAACL 2024, and COLING-LREC 2024 (co-first author)
  • 2023: Two papers at ACL 2023 (one first-author); first-author paper at AAAI 2023
  • 2022: First-author paper at ACL 2022
  • 2021: First-author paper at ICML 2021 Workshop
  • Led or contributed to key project releases: Dolphin (multimodal document parsing model), WildDoc (first benchmark for document understanding in natural environments), VORA (challenging encoder-based MLLM architectures), LayTextLLM (interleaving layout and text for LLMs, achieving SOTA on multiple Document AI benchmarks), MTVQA (first text-centric multilingual multimodal benchmark for LLMs)