4 papers · 1 filter
Next-Token Prediction Task Assumes Optimal Data Ordering for LLM Training in Proof Generation
Chenyang An, Shima Imani, Feng Yao +8
In the field of large language model (LLM)-based proof generation, despite extensive training on large datasets such as ArXiv, LLMs still exhibit only modest performance on proving…
The Price of Format: Diversity Collapse in LLMs
Longfei Yun, Chenyang An, Zilong Wang +2
Instruction-tuned large language models (LLMs) employ structured templates, such as role markers and special tokens, to enforce format consistency during inference. However, we ide…
Linear Correlation in LM's Compositional Generalization and Hallucination
Letian Peng, Chenyang An, Shibo Hao +2
The generalization of language models (LMs) is undergoing active debates, contrasting their potential for general intelligence with their struggles with basic knowledge composition…
Correlation and Navigation in the Vocabulary Key Representation Space of Language Models
Letian Peng, Chenyang An, Jingbo Shang
Language model (LM) decoding is based on the next-token prediction (NTP) probability distribution. For neural LMs (e.g., Transformer-based), NTP distribution is essentially a softm…