Comparing GPT-2 and Llama-2 on belief-tracking story prompts, the paper finds that added context and higher temperature reduce the probability of each model's own most likely next token.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Exploring Next Token Prediction in Theory of Mind (ToM) Tasks: Comparative Experiments with GPT-2 and LLaMA-2 AI Models
Comparing GPT-2 and Llama-2 on belief-tracking story prompts, the paper finds that added context and higher temperature reduce the probability of each model's own most likely next token.