PRISM: A Dual View of LLM Reasoning through Semantic Flow and Latent Computation

ArXi:2603.22754v1 Announce Type: new Large language models (LLMs) solve complex problems by generating multi-step reasoning traces. Yet these traces are typically analyzed from only one of two perspectives: the sequence of tokens across different reasoning steps in the generated text, or the hidden-state vectors across model layers within one step. We