6 papers
Higher-Dimensional Rotary Position Embedding
Yixing Li, Ruobing Xie, Yudong Zhang +3
Transformers rely on position embedding mechanisms in long context modeling in most cases. Rotary Position Embedding (RoPE) embeds positional information with independent 2D rotati…
VAD: Attributing Visual Evidence for Target Reconstruction in Multimodal On-Policy Distillation
Kangning Zhang, Yixing Li, Shuai Shao +9
Multimodal on-policy distillation (OPD) transfers fine-grained visual knowledge by supervising student-generated trajectories with a privileged-view teacher. Yet its next-token cor…
TransMamba: A Sequence-Level Hybrid Transformer-Mamba Language Model
Yixing Li, Ruobing Xie, Zhen Yang +8
Transformers are the cornerstone of modern large language models, but their quadratic computational complexity limits efficiency in long-sequence processing. Recent advancements in…
Scaling Laws for Floating Point Quantization Training
Xingwu Sun, Shuaipeng Li, Ruobing Xie +13
Low-precision training is considered an effective strategy for reducing both training and downstream inference costs. Previous scaling laws for precision mainly focus on integer qu…
Continuous Speech Tokenizer in Text To Speech
Yixing Li, Ruobing Xie, Xingwu Sun +2
The fusion of speech and language in the era of large language models has garnered significant attention. Discrete speech token is often utilized in text-to-speech tasks for speech…
Direct Preference Knowledge Distillation for Large Language Models
Yixing Li, Yuxian Gu, Li Dong +3
In the field of large language models (LLMs), Knowledge Distillation (KD) is a critical technique for transferring capabilities from teacher models to student models. However, exis…