Partially replacing chain-of-thought tokens with discrete latent tokens during fine-tuning improves LLM reasoning accuracy and reduces generated token count.
Theoremqa: A theorem-driven question answering dataset
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Token Assorted: Mixing Latent and Text Tokens for Improved Language Model Reasoning
Partially replacing chain-of-thought tokens with discrete latent tokens during fine-tuning improves LLM reasoning accuracy and reduces generated token count.