Saved in:
Bibliographic Details
Main Authors: Song, Ruizhuo, Yuan, Beiming
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.15387
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911255217831936
author Song, Ruizhuo
Yuan, Beiming
author_facet Song, Ruizhuo
Yuan, Beiming
contents Despite deep learning's broad success, its abstract-reasoning bottleneck persists. We tackle Raven's Progressive Matrices (RPM), the benchmark for pattern, reasoning and problem-solving intelligence. We model the full causal chain image $\rightarrow$ attributes $\rightarrow$ progressive patterns $\rightarrow$ consistency $\rightarrow$ answer and build the baseline DIO. Yet DIO's mutual-information lower-bound objective does not embed human logic: the bound is loose and statistic-based, ignoring causal subject-object links. We therefore present three refinements. 1) Brando introduces trainable negative options to tighten the variational bound. 2) WORLD replaces generation with a Gaussian-mixture feature model that supplies infinite, weighted negatives, further tightening the bound. 3) DIEGO adds metadata supervision to rectify the "attributes $\rightarrow$ patterns" semantic gap, aligning representations with human rules. These upgrades substantially boost discriminative RPM accuracy and, for the first time, let DIO generate valid answers in open-ended RPM. The work provides causal-driven design guidelines, objective-refinement strategies and cross-modal insights for abstract-reasoning research.
format Preprint
id arxiv_https___arxiv_org_abs_2508_15387
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DIO: Refining Mutual Information and Causal Chain to Enhance Machine Abstract Reasoning Ability
Song, Ruizhuo
Yuan, Beiming
Computer Vision and Pattern Recognition
Despite deep learning's broad success, its abstract-reasoning bottleneck persists. We tackle Raven's Progressive Matrices (RPM), the benchmark for pattern, reasoning and problem-solving intelligence. We model the full causal chain image $\rightarrow$ attributes $\rightarrow$ progressive patterns $\rightarrow$ consistency $\rightarrow$ answer and build the baseline DIO. Yet DIO's mutual-information lower-bound objective does not embed human logic: the bound is loose and statistic-based, ignoring causal subject-object links. We therefore present three refinements. 1) Brando introduces trainable negative options to tighten the variational bound. 2) WORLD replaces generation with a Gaussian-mixture feature model that supplies infinite, weighted negatives, further tightening the bound. 3) DIEGO adds metadata supervision to rectify the "attributes $\rightarrow$ patterns" semantic gap, aligning representations with human rules. These upgrades substantially boost discriminative RPM accuracy and, for the first time, let DIO generate valid answers in open-ended RPM. The work provides causal-driven design guidelines, objective-refinement strategies and cross-modal insights for abstract-reasoning research.
title DIO: Refining Mutual Information and Causal Chain to Enhance Machine Abstract Reasoning Ability
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.15387