An LLM-native five-factor psychometric instrument shows self-reports fail to predict behavior even on constructs derived from LLM behavior, and LLM judges share a variance source humans do not.
Tuhin Chakrabarty, Philippe Laban, and Chien-Sheng Wu
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
other 1
citation-polarity summary
years
2026 3roles
other 1polarities
unclear 1representative citing papers
A per-token feature from temperature-induced changes in LLM token distributions predicts within-prompt creativity rank at Spearman rho 0.918 vs LLM judges and 0.870 vs humans, outperforming perplexity, entropy, top-1 margin, and compression baselines.
StoryReward, trained on a new 100k story preference dataset, sets state-of-the-art performance on the introduced StoryRMB benchmark for aligning LLM stories with human preferences.
citing papers explorer
-
An LLM-Native Psychometric Instrument Reveals a Self-Report--Behavior Gap Across 25 Models
An LLM-native five-factor psychometric instrument shows self-reports fail to predict behavior even on constructs derived from LLM behavior, and LLM judges share a variance source humans do not.
-
Before and After Temperature: A Distributional View of Creative LLM Generation
A per-token feature from temperature-induced changes in LLM token distributions predicts within-prompt creativity rank at Spearman rho 0.918 vs LLM judges and 0.870 vs humans, outperforming perplexity, entropy, top-1 margin, and compression baselines.
-
StoryAlign: Evaluating and Training Reward Models for Story Generation
StoryReward, trained on a new 100k story preference dataset, sets state-of-the-art performance on the introduced StoryRMB benchmark for aligning LLM stories with human preferences.