Papers by Wenfeng Xie
DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models (2024.acl-long)
Copied to clipboard
Damai Dai, Chengqi Deng, Chenggang Zhao, R.x. Xu, Huazuo Gao, Deli Chen, Jiashi Li, Wangding Zeng, Xingkai Yu, Y. Wu, Zhenda Xie, Y.k. Li, Panpan Huang, Fuli Luo, Chong Ruan, Zhifang Sui, Wenfeng Liang
| Challenge: | Mixture-of-Experts (MoE) architectures face challenges in ensuring expert specialization . despite the promising performance, scaling language models to an extremely large scale is associated with exceedingly high computational costs. |
| Approach: | They propose an architecture that allows for ultimate expert specialization by segmenting experts into mN ones and activating mK from them. |
| Outcome: | The proposed architecture achieves comparable performance with GShard with 2B parameters and computation. |
Mitigating Boundary Ambiguity and Inherent Bias for Text Classification in the Era of Large Language Models (2024.findings-acl)
Copied to clipboard
| Challenge: | a new text classification framework for large language models addresses the problem of boundary ambiguity and inherent biases in LLMs. |
| Approach: | They propose a two-stage classification framework for large language models to mitigate bottlenecks . their approach uses pairwise comparisons to efficiently narrow down options . |
| Outcome: | The proposed framework reduces the number of options and improves on four datasets. |
Reinforcement Learning with Token-level Feedback for Controllable Text Generation (2024.findings-naacl)
Copied to clipboard
| Challenge: | Existing methods for controllable text generation are guided by coarse-grained feedback, which may lead to suboptimal performance owing to semantic twists or progressions within sentences. |
| Approach: | They propose a reinforcement learning algorithm which formulates TOken-LEvel rewards for controllable text generation and employs a "first-quantize-then-noise" paradigm to enhance the robustness of the RL algorithm. |
| Outcome: | The proposed algorithm can achieve superior performance on single-attribute and multi-attract control tasks. |
Modeling Historical Relevant and Local Frequency Context for Representation-Based Temporal Knowledge Graph Forecasting (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing representation-based approaches neglect candidate-specific temporal context, resulting in serious information loss or homogeneous prediction. |
| Approach: | They propose a temporal representation learning model that incorporates temporal contexts of candidates and models temporal contextual information from historiCal Relevant context and locAl Frequency contexT. |
| Outcome: | The proposed model can leverage temporal contextual information to achieve differential predictions on six benchmark datasets. |
Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention (2025.acl-long)
Copied to clipboard
Jingyang Yuan, Huazuo Gao, Damai Dai, Junyu Luo, Liang Zhao, Zhengyan Zhang, Zhenda Xie, Yuxing Wei, Lean Wang, Zhiping Xiao, Yuqing Wang, Chong Ruan, Ming Zhang, Wenfeng Liang, Wangding Zeng
| Challenge: | Long-context modeling is crucial for next-generation language models, but high computational cost of standard attention mechanisms poses significant computational challenges. |
| Approach: | They propose a natively trained Sparse Attention mechanism that integrates algorithms with hardware-aligned optimizations to achieve efficient long-context modeling. |
| Outcome: | The proposed model maintains or exceeds Full Attention models across general benchmarks, long-context tasks, and instruction-based reasoning. |
TREA: Tree-Structure Reasoning Schema for Conversational Recommendation (2023.acl-long)
Copied to clipboard
| Challenge: | Recent reasoning-based models cannot fully figure out complex causal relationships between mentioned entities with external knowledge. |
| Approach: | They propose a Tree structure Reasoning schEmA that constructs a multi-hierarchical scalable tree as the reasoning structure to clarify the causal relationships between mentioned entities. |
| Outcome: | Extensive experiments on two public CRS datasets show the proposed model works. |
Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models (2026.acl-long)
Copied to clipboard
Xin Cheng, Wangding Zeng, Damai Dai, Qinyu Chen, Bingxuan Wang, Zhenda Xie, Kezhao Huang, Xingkai Yu, Zhewen Hao, Han Zhang, Yu-Kun Li, Huishuai Zhang, Dongyan Zhao, Wenfeng Liang
| Challenge: | Mixture-of-Experts (MoE) scales capacity via conditional computation, but lacks knowledge lookup primitive. |
| Approach: | They propose a conditional memory instantiated via Deep Sparse Embedding (DSE) they propose 'u-shaped scaling law' that identifies optimal balance between MoE experts and DSE memory . |
| Outcome: | The proposed model outperforms an iso-parameter and isoFLOPs MoE baseline across knowledge and reasoning benchmarks and is infrastructure-efficient. |