paper-with-me

홈 › Papers

Supervising the Transfer of Reasoning Patterns in VQA

2021-06-10 · NeurIPS 2021 12 · Corentin Kervadec, Christian Wolf, Grigory Antipov, Moez Baccouche, Madiha Nadri

Methods for Visual Question Anwering (VQA) are notorious for leveraging dataset biases rather than performing reasoning, hindering generalization. It has been recently shown that better reasoning patterns emerge in attention layers of a state-of-the-art VQA model when they are trained on perfect (oracle) visual inputs. This provides evidence that deep neural networks can learn to reason when training conditions are favorable enough. However, transferring this learned knowledge to deployable models is a challenge, as much of it is lost during the transfer. We propose a method for knowledge transfer based on a regularization term in our loss function, supervising the sequence of required reasoning operations. We provide a theoretical analysis based on PAC-learning, showing that such program prediction can lead to decreased sample complexity under mild hypotheses. We also demonstrate the effectiveness of this approach experimentally on the GQA dataset and show its complementarity to BERT-like self-supervised pre-training.

📄 PDF Abstract BibTeX arXiv:2106.05597

Code (0)

등록된 구현이 없습니다.

Tasks

PAC learningTransfer LearningVisual Question Answering (VQA)

Similar Papers 제목 키워드 기반

Learning to Think Like a Cartoon Captionist: Incongruity-Resolution Supervision for Multimodal Humor Understanding

2026-04-16 · Hatice Merve Vural, Doga Kukul, Ege Erdem Ozlu, Demir Ekin Arikan 외 arxiv

Humor is one of the few cognitive tasks where getting the reasoning right matters as much as getting the answer right. While recent work evaluates humor understanding on benchmarks such as the New Yorker Cartoon Caption …

TaTToo: Tool-Grounded Thinking PRM for Test-Time Scaling in Tabular Reasoning

2025-10-07 · Jiaru Zou, Soumya Roy, Vinay Kumar Verma, Ziyi Wang 외 arxiv

Process Reward Models (PRMs) have recently emerged as a powerful framework for enhancing the reasoning capabilities of large reasoning models (LRMs), particularly in the context of test-time scaling (TTS). However, their…

Reinforcement LearningTable Retrieval

Every Coin Has Two Sides: On the Dual Nature of Generalization in On-Policy Distillation of Large Language Models

2026-08-17 · Zhaoyi Li, Deyang Kong, Yuan Wei, Evan Yang 외 arxiv

On-policy distillation (OPD) transfers teacher capabilities by supervising trajectories sampled from the student's own policy, yet its generalization behavior remains poorly understood, as most studies evaluate OPD on a …

Supervised Convex Clustering

2020-05-25 · Minjie Wang, Tianyi Yao, Genevera I. Allen

Clustering has long been a popular unsupervised learning approach to identify groups of similar objects and discover patterns from unlabeled data in many applications. Yet, coming up with meaningful interpretations of th…

ClusteringDiagnostic

OPLD: On-Policy Latent Distillation for Multimodal Reasoning

2026-07-30 · Shoutai Zhu, Tianyang Xu, Bin Sun, Mingyuan Xu 외 arxiv

Interleaved multimodal Chain-of-Thought (CoT) improves visual reasoning by incorporating auxiliary visual evidence into intermediate reasoning. However, existing approaches remain constrained by externally defined reason…

Multimodal ReasoningVisual Reasoning