conftrace_

logit flow

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

back attention (1) reasoning accuracy (1) latent reasoning (1) transformer layer (1) interpretability method (1) multi-hop reasoning (1) hidden state (1) attention mechanism (1)

Papers

Back Attention: Understanding and Enhancing Multi-Hop Reasoning in Large Language Models EMNLP 2025