System 2 Reasoning for Human-AI Alignment: Generality and Adaptivity via ARC-AGI
Fuente:
arXiv
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915442650513408 |
|---|---|
| author | Kim, Sejin Kim, Sundong |
| author_facet | Kim, Sejin Kim, Sundong |
| contents | Despite their broad applicability, transformer-based models still fall short in System~2 reasoning, lacking the generality and adaptivity needed for human--AI alignment. We examine weaknesses on ARC-AGI tasks, revealing gaps in compositional generalization and novel-rule adaptation, and argue that closing these gaps requires overhauling the reasoning pipeline and its evaluation. We propose three research axes: (1) Symbolic representation pipeline for compositional generality, (2) Interactive feedback-driven reasoning loop for adaptivity, and (3) Test-time task augmentation balancing both qualities. Finally, we demonstrate how ARC-AGI's evaluation suite can be adapted to track progress in symbolic generality, feedback-driven adaptivity, and task-level robustness, thereby guiding future work on robust human--AI alignment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_07866 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | System 2 Reasoning for Human-AI Alignment: Generality and Adaptivity via ARC-AGI Kim, Sejin Kim, Sundong Artificial Intelligence Despite their broad applicability, transformer-based models still fall short in System~2 reasoning, lacking the generality and adaptivity needed for human--AI alignment. We examine weaknesses on ARC-AGI tasks, revealing gaps in compositional generalization and novel-rule adaptation, and argue that closing these gaps requires overhauling the reasoning pipeline and its evaluation. We propose three research axes: (1) Symbolic representation pipeline for compositional generality, (2) Interactive feedback-driven reasoning loop for adaptivity, and (3) Test-time task augmentation balancing both qualities. Finally, we demonstrate how ARC-AGI's evaluation suite can be adapted to track progress in symbolic generality, feedback-driven adaptivity, and task-level robustness, thereby guiding future work on robust human--AI alignment. |
| title | System 2 Reasoning for Human-AI Alignment: Generality and Adaptivity via ARC-AGI |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2410.07866 |