9 citations · 26 across the 51 of their papers we have counts for
8 papers · 1 filter
PhaseQuant: Mitigating Activation Quantization Collapse via Residual-Stream Dynamics
Zimo Zhao, Maolin Wang, Bowen Yu +3
Post-training quantization (PTQ) is essential for efficient large language model (LLM) inference, but reliably quantizing activations remains challenging when weights, activations,…
BRIDGE: Bridging Reasoning In Distillation Gap Elimination via Structure-Aware Masking
Bowen Yu, Sheng Zhang, Binhao Wang +8
Chain-of-Thought (CoT) reasoning has significantly improved LLMs' mathematical problem-solving capabilities, but distilling such capabilities into smaller models remains challengin…
DANCE: Resource-Efficient Neural Architecture Search with Data-Aware and Continuous Adaptation
Maolin Wang, Tianshuo Wei, Sheng Zhang +6
Neural Architecture Search (NAS) has emerged as a powerful approach for automating neural network design. However, existing NAS methods face critical limitations in real-world depl…
Reasoning through Exploration: A Reinforcement Learning Framework for Robust Function Calling
Bingguang Hao, Zengzhuang Xu, Maolin Wang +9
The effective training of Large Language Models (LLMs) for function calling faces a critical challenge: balancing exploration of complex reasoning paths with stable policy optimiza…
FunReason: Enhancing Large Language Models' Function Calling via Self-Refinement Multiscale Loss and Automated Data Refinement
Bingguang Hao, ZengZhuang Xu, Maolin Wang +9
The integration of large language models (LLMs) with function calling has emerged as a crucial capability for enhancing their practical utility in real-world applications. However,…
MetaLoRA: Tensor-Enhanced Adaptive Low-Rank Fine-tuning
Maolin Wang, Xiangyu Zhao
There has been a significant increase in the deployment of neural network models, presenting substantial challenges in model adaptation and fine-tuning. Efficient adaptation is cru…
Cumulative Distribution Function based General Temporal Point Processes
Maolin Wang, Yu Pan, Zenglin Xu +6
Temporal Point Processes (TPPs) hold a pivotal role in modeling event sequences across diverse domains, including social networking and e-commerce, and have significantly contribut…
Federated Knowledge Graph Completion via Latent Embedding Sharing and Tensor Factorization
Maolin Wang, Dun Zeng, Zenglin Xu +2
Knowledge graphs (KGs), which consist of triples, are inherently incomplete and always require completion procedure to predict missing triples. In real-world scenarios, KGs are dis…