Profiles2024-2025
Junshu Sun
University of Chinese Academy of Sciences
8Papers
0Last author
10Unique co-authors
2First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers8
- VL-SAE: Interpreting and Enhancing Vision-Language Alignment with a Unified Concept SetAuthor 2 of 4NeurIPS 2025 · 13 citations
- Edit Less, Achieve More: Dynamic Sparse Neuron Masking for Lifelong Knowledge Editing in LLMsAuthor 2 of 5NeurIPS 2025 · 8 citations
- Relieving the Over-Aggregating Effect in Graph TransformersAuthor 1 of 5NeurIPS 2025 · 3 citations
- Enhancing Pre-trained Representation Classifiability can Boost its InterpretabilityAuthor 3 of 6ICLR 2025
- Learning Fine-Grained Representations through Textual Token Disentanglement in Composed Video RetrievalAuthor 4 of 6ICLR 2025
- Video Language Model Pretraining with Spatio-temporal MaskingAuthor 3 of 6CVPR 2025
- Towards Dynamic Message Passing on GraphsAuthor 1 of 5NeurIPS 2024 · 19 citations
- Expanding Sparse Tuning for Low Memory UsageAuthor 2 of 5NeurIPS 2024 · 12 citations
