Addressing and Visualizing Misalignments in Human Task-Solving Trajectories

TL;DR

This paper formalizes types of misalignments in human task trajectories, proposes detection and intention estimation methods, and shows that aligning AI training with human trajectories improves reasoning mimicry.

Contribution

It introduces a formal framework for misalignments, a heuristic detection algorithm, and an intention estimation method to enhance AI's mimicry of human reasoning.

Findings

Misalignments can be categorized into three types.

Trajectory alignment improves AI performance in mimicking human reasoning.

Intention alignment is crucial for effective AI training.

Abstract

Understanding misalignments in human task-solving trajectories is crucial for enhancing AI models trained to closely mimic human reasoning. This study categorizes such misalignments into three types: (1) lack of functions to express intent, (2) inefficient action sequences, and (3) incorrect intentions that cannot solve the task. To address these issues, we first formalize and define these three misalignment types in a unified framework. We then propose a heuristic algorithm to detect misalignments in ARCTraj trajectories and analyze their impact hierarchically and quantitatively. We also present an intention estimation method based on our formalism that infers missing alignment between user actions and intentions. Through trajectory alignment, we experimentally demonstrate that AI models trained on human task-solving trajectories improve performance in mimicking human reasoning. Based on hierarchical analysis and experiments, we highlight the importance of trajectory-intention alignment and demonstrate the effectiveness of intention-aligned training.