activity
20242026
collaborators

5 papers

cs.CL2026

A Unified View of Attention and Residual Sinks: Outlier-Driven Rescaling is Essential for Transformer Training

Zihan Qiu, Zeyu Huang, Kaiyue Wen +16

We investigate the functional role of emergent outliers in large language models, specifically attention sinks (a few tokens that consistently receive large attention logits) and r…

cs.CL2025

A Controllable Examination for Long-Context Language Models

Yijun Yang, Zeyu Huang, Wenhao Zhu +4

Existing frameworks for evaluating long-context language models (LCLM) can be broadly categorized into real-world applications (e.g, document summarization) and synthetic tasks (e.…

cs.CL2025

Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free

Zihan Qiu, Zekun Wang, Bo Zheng +10

Gating mechanisms have been widely utilized, from early models like LSTMs and Highway Networks to recent state space models, linear attention, and also softmax attention. Yet, exis…

cs.LG2025

Demons in the Detail: On Implementing Load Balancing Loss for Training Specialized Mixture-of-Expert Models

Zihan Qiu, Zeyu Huang, Bo Zheng +7

This paper revisits the implementation of oad-alancing oss (LBL) when training Mixture-of-Experts (MoEs) models. Specifically, LBL for MoEs is d…

cs.AI2024

Post-hoc Reward Calibration: A Case Study on Length Bias

Zeyu Huang, Zihan Qiu, Zili Wang +2

Reinforcement Learning from Human Feedback aligns the outputs of Large Language Models with human values and preferences. Central to this process is the reward model (RM), which tr…