Profiles2022-2025
Damai Dai
MOE Key Lab of Computational Linguistics, School of Computer Science, Peking University
9Papers
0Last author
19Unique co-authors
4First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers9
- Exploring Activation Patterns of Parameters in Language ModelsAuthor 2 of 3AAAI 2025
- DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language ModelsAuthor 1 of 17ACL 2024 · 171 citations
- Why Can GPT Learn In-Context? Language Models Secretly Perform Gradient Descent as Meta-OptimizersAuthor 1 of 7ACL Findings 2023 · 154 citations
- Label Words are Anchors: An Information Flow Perspective for Understanding In-Context LearningAuthor 3 of 8EMNLP 2023 · 26 citations
- Denoising Bottleneck with Mutual Information Maximization for Video Multimodal FusionAuthor 2 of 7ACL 2023 · 11 citations
- Bi-Drop: Enhancing Fine-tuning Generalization via Synchronous sub-net Estimation and OptimizationAuthor 3 of 8EMNLP Findings 2023
- Calibrating Factual Knowledge in Pretrained Language ModelsAuthor 2 of 6EMNLP Findings 2022 · 39 citations
- Knowledge Neurons in Pretrained TransformersAuthor 1 of 6ACL 2022
- StableMoE: Stable Routing Strategy for Mixture of ExpertsAuthor 1 of 7ACL 2022
