VQ-Logits: Compressing the Output Bottleneck of Large Language Models via Vector Quantized Logits
Paper
•
2505.10202
•
Published
shaojintian
shaojintian
AI & ML interests
None yet
Organizations
authored
4
papers
3 months ago
Power-Law Decay Loss for Large Language Model Finetuning: A Theory Perspective
Paper
•
2505.16900
•
Published
ComplexFormer: Disruptively Advancing Transformer Inference Ability via Head-Specific Complex Vector Attention
Paper
•
2505.10222
•
Published
Towards Analyzing and Understanding the Limitations of VAPO: A Theoretical Perspective
Paper
•
2505.17997
•
Published