Profiles2020-2026
Hua Wu
Baidu Inc., Beijing
26Papers
4Last author
71Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers26
- ProxyAttn: Guided Sparse Attention via Representative HeadsAuthor 4 of 7ICLR 2026 · 9 citations
- Enough is as good as a feast: A Comprehensive Analysis of How Reinforcement Learning Mitigates Task Conflicts in LLMsAuthor 6 of 8ICLR 2026 · 2 citations
- Sparse Growing Transformer: Training-Time Sparse Depth Allocation via Progressive Attention LoopingAuthor 10 of 12ACL Findings 2026
- Uncertainty-Aware Routing for Principled Alignment with MoE DynamicsAuthor 10 of 11ACL 2026
- NACL: A General and Effective KV Cache Eviction Framework for LLM at Inference TimeAuthor 10 of 10ACL 2024 · 8 citations
- LEMON: Reviving Stronger and Smaller LMs from Larger LMs with Linear Parameter FusionAuthor 8 of 8ACL 2024 · 1 citation
- Universal Information Extraction as Unified Semantic MatchingAuthor 8 of 8AAAI 2023 · 96 citations
- ERNIE-ViLG 2.0: Improving Text-to-Image Diffusion Model with Knowledge-Enhanced Mixture-of-Denoising-ExpertsAuthor 14 of 15CVPR 2023
- ERNIE-ViL: Knowledge Enhanced Vision-Language Representations through Scene GraphsAuthor 6 of 7AAAI 2021 · 414 citations
- PLATO-2: Towards Building an Open-Domain Chatbot via Curriculum LearningAuthor 4 of 9ACL Findings 2021 · 94 citations
- Correcting Chinese Spelling Errors with Phonetic Pre-trainingAuthor 7 of 8ACL Findings 2021 · 67 citations
- BASS: Boosting Abstractive Summarization with Unified Semantic GraphAuthor 7 of 8ACL 2021
- Discovering Dialog Structure Graph for Coherent Dialog GenerationAuthor 5 of 6ACL 2021
- ERNIE-Doc: A Retrospective Long-Document Modeling TransformerAuthor 6 of 7ACL 2021
- UNIMO: Towards Unified-Modal Understanding and Generation via Cross-Modal Contrastive LearningAuthor 7 of 8ACL 2021
- ERNIE 2.0: A Continual Pre-Training Framework for Language UnderstandingAuthor 6 of 7AAAI 2020 · 885 citations
- SKEP: Sentiment Knowledge Enhanced Pre-training for Sentiment AnalysisAuthor 6 of 8ACL 2020 · 265 citations
- PLATO: Pre-trained Dialogue Generation Model with Discrete Latent VariableAuthor 4 of 5ACL 2020 · 229 citations
- Towards Conversational Recommendation over Multi-Type DialogsAuthor 4 of 6ACL 2020 · 157 citations
- Leveraging Graph to Improve Abstractive Multi-Document SummarizationAuthor 4 of 6ACL 2020 · 118 citations
- Synchronous Speech Recognition and Speech-to-Text Translation with Interactive DecodingAuthor 6 of 8AAAI 2020 · 73 citations
- Knowledge Graph Grounded Goal Planning for Open-Domain Conversation GenerationAuthor 4 of 5AAAI 2020 · 69 citations
- Conversational Graph Grounded Policy Learning for Open-Domain Conversation GenerationAuthor 4 of 6ACL 2020 · 52 citations
- Learning Adaptive Segmentation Policy for Simultaneous TranslationAuthor 4 of 5EMNLP 2020 · 41 citations
- Diversified Multiple Instance Learning for Document-Level Multi-Aspect Sentiment ClassificationAuthor 5 of 6EMNLP 2020 · 22 citations
- Syntactic and Semantic-driven Learning for Open Information ExtractionAuthor 7 of 7EMNLP Findings 2020 · 6 citations
