# ActTraitBench: Quantifying the Knowledge-Decision Gap in Large Language Models via Human-Grounded Behavioral Validation

Source: [ActTraitBench: Quantifying the Knowledge-Decision Gap in Large Language Models via Human-Grounded Behavioral Validation](<https://www.sota2.com/research/paper/6a191dd4dbd7150506f82ad7>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-30T05:00:11.986Z

Authors: Yutong Yang, Chenxi Miao, Weikang Li, Yunfang Wu

Published: 2026-05-28T11:40:35.000Z

## Abstract

While Large Language Models (LLMs) can convincingly simulate personas in explicit self-reports, they often deviate in implicit behavioral decisions, revealing a substantial Knowledge-Decision Gap ($G\_{\\text{KD}}$). Existing benchmarks struggle to measure this asymmetry due to limited construct validity, multi-dimensional entanglement, and distributional biases in LLM-based evaluation. To address these issues, we propose ActTraitBench, a human-grounded evaluation framework for measuring personality consistency in LLMs. Grounded in empirical human data, ActTraitBench establishes one-to-one mappings between psychometric facets and behavioral paradigms, and applies a Distributional Calibration via Quantile Mapping procedure to align LLM-judge score distributions with human norms. Experiments on 14 mainstream LLMs reveal a pervasive knowledge-decision asymmetry, where larger and more capable models often exhibit stronger behavioral divergence despite highly consistent self-reports. To mitigate this gap, we further introduce the Chain of Cognitive Alignment (CoCA), a plug-and-play inference-time intervention that improves alignment in reasoning-capable frontier models while exposing clear capability limitations in smaller architectures.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Personality Alignment Evaluation on Big Five Inventory (BFI) and Behavioral Tasks](<https://www.sota2.com/research/sota/personality-alignment-evaluation-on-big-five-inventory-bfi-and-behavioral-tasks.md>): Personality Alignment Evaluation; Big Five Inventory (BFI) and Behavioral Tasks; Knowledge Score (E): not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2605.29791>)

[PDF](<https://arxiv.org/pdf/2605.29791.pdf>)
