From Hindsight to Foresight: Self-Encouraged Hindsight Distillation for Knowledge-based Visual Question Answering

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhao, Yu, Zhang, Ying, Sui, Xuhui, Zhou, Baohang, Shen, Li, Tao, Dacheng
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911559183237120
author Zhao, Yu
Zhang, Ying
Sui, Xuhui
Zhou, Baohang
Shen, Li
Tao, Dacheng
author_facet Zhao, Yu
Zhang, Ying
Sui, Xuhui
Zhou, Baohang
Shen, Li
Tao, Dacheng
contents Knowledge-based Visual Question Answering (KBVQA) necessitates external knowledge incorporation beyond cross-modal understanding. Existing KBVQA methods either utilize implicit knowledge in multimodal large language models (MLLMs) via in-context learning or explicit knowledge via retrieval augmented generation. However, their reasoning processes remain implicit, without explicit multi-step trajectories from MLLMs. To address this gap, we provide a Hindsight Distilled Reasoning (HinD) framework with Knowledge Encouragement Preference Optimization, aiming at self-encouraging the knowledge reasoning ability inside the MLLM. First, we construct the Hindsight Teacher by prompting the MLLM to complete the reasoning process with knowing the right answer, obtaining Hindsight-Zero training data. Then, the Foresight Student, without knowing the answer, learns the golden trajectories from Hindsight: (1) Hindsight Distillation Fine-Tuning (HDFT) to self-distill the Hindsight-Zero into a modularized Chain-of-Thought (CoT) Generator and a Knowledge Generator for sequential steps and discrete facts generation, respectively; (2) Knowledge Encouragement Preference Optimization (KEPO) to encourage the under-confident but relevant knowledge inside the MLLM and suppress the over-confident but irrelevant one. Experiments on OK-VQA and A-OKVQA validate the effectiveness of HinD, showing that HinD with 7-8B MLLM achieves superior performance without commercial model APIs or retrieved knowledge.
format Preprint
id arxiv_https___arxiv_org_abs_2511_11132
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle From Hindsight to Foresight: Self-Encouraged Hindsight Distillation for Knowledge-based Visual Question Answering
Zhao, Yu
Zhang, Ying
Sui, Xuhui
Zhou, Baohang
Shen, Li
Tao, Dacheng
Computer Vision and Pattern Recognition
Knowledge-based Visual Question Answering (KBVQA) necessitates external knowledge incorporation beyond cross-modal understanding. Existing KBVQA methods either utilize implicit knowledge in multimodal large language models (MLLMs) via in-context learning or explicit knowledge via retrieval augmented generation. However, their reasoning processes remain implicit, without explicit multi-step trajectories from MLLMs. To address this gap, we provide a Hindsight Distilled Reasoning (HinD) framework with Knowledge Encouragement Preference Optimization, aiming at self-encouraging the knowledge reasoning ability inside the MLLM. First, we construct the Hindsight Teacher by prompting the MLLM to complete the reasoning process with knowing the right answer, obtaining Hindsight-Zero training data. Then, the Foresight Student, without knowing the answer, learns the golden trajectories from Hindsight: (1) Hindsight Distillation Fine-Tuning (HDFT) to self-distill the Hindsight-Zero into a modularized Chain-of-Thought (CoT) Generator and a Knowledge Generator for sequential steps and discrete facts generation, respectively; (2) Knowledge Encouragement Preference Optimization (KEPO) to encourage the under-confident but relevant knowledge inside the MLLM and suppress the over-confident but irrelevant one. Experiments on OK-VQA and A-OKVQA validate the effectiveness of HinD, showing that HinD with 7-8B MLLM achieves superior performance without commercial model APIs or retrieved knowledge.
title From Hindsight to Foresight: Self-Encouraged Hindsight Distillation for Knowledge-based Visual Question Answering
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.11132