paper-with-me

Papers

Difficult Task Yes but Simple Task No: Unveiling the Laziness in Multimodal LLMs

2024-10-15 · Sihang Zhao, Youliang Yuan, Xiaoying Tang, Pinjia He

Multimodal Large Language Models (MLLMs) demonstrate a strong understanding of the real world and can even handle complex tasks. However, they still fail on some straightforward visual question-answering (VQA) problems. This paper dives deeper into this issue, revealing that models tend to err when answering easy questions (e.g. Yes/No questions) about an image, even though they can correctly describe it. We refer to this model behavior discrepancy between difficult and simple questions as model laziness. To systematically investigate model laziness, we manually construct LazyBench, a benchmark that includes Yes/No, multiple choice, short answer questions, and image description tasks that are related to the same subjects in the images. Based on LazyBench, we observe that laziness widely exists in current advanced MLLMs (e.g. GPT-4o, Gemini-1.5-pro, Claude 3 and LLaVA-v1.5-13B), and it is more pronounced on stronger models. We also analyze the VQA v2 (LLaVA-v1.5-13B) benchmark and find that about half of its failure cases are caused by model laziness, which further highlights the importance of ensuring that the model fully utilizes its capability. To this end, we conduct preliminary exploration on how to mitigate laziness and find that chain of thought (CoT) can effectively address this issue.

📄 PDF Abstract BibTeX arXiv:2410.11437

Code (1)

Akutagawa1998/LazyBench 공식 구현

Tasks

Image DescriptionMultiple-choiceQuestion AnsweringVisual Question AnsweringVisual Question Answering (VQA)

Similar Papers 제목 키워드 기반

UniCat: Crafting a Stronger Fusion Baseline for Multimodal Re-Identification

2023-10-28 · Jennifer Crawford, Haoli Yin, Luke McDermott, Daniel Cummings

Multimodal Re-Identification (ReID) is a popular retrieval task that aims to re-identify objects across diverse data streams, prompting many researchers to integrate multiple modalities into a unified representation. Whi…

Retrieval

Quantifying Laziness, Decoding Suboptimality, and Context Degradation in Large Language Models

2025-12-19 · Yiqing Ma, Jung-Hua Liu arxiv

Large Language Models (LLMs) often exhibit behavioral artifacts such as laziness (premature truncation of responses or partial compliance with multi-part requests), decoding suboptimality (failure to select higher-qualit…

Ethically Aligned Opportunistic Scheduling for Productive Laziness

2019-01-02 · Han Yu, Chunyan Miao, Yongqing Zheng, Lizhen Cui 외

In artificial intelligence (AI) mediated workforce management systems (e.g., crowdsourcing), long-term success depends on workers accomplishing tasks productively and resting well. This dual objective can be summarized b…

ManagementScheduling

Modality Laziness: Everybody's Business is Nobody's Business

2021-09-29 · Chenzhuang Du, Jiaye Teng, Tingle Li, Yichen Liu 외

Models fusing multiple modalities receive more information and can outperform their uni-modal counterparts. However, existing multi-modal training approaches often suffer from learning insufficient representations of eac…

Partially Lazy Gradient Descent for Smoothed Online Learning

2026-01-22 · Naram Mhaisen, George Iosifidis arxiv

We introduce \textsc{$k$-lazyGD}, an online learning algorithm that bridges the gap between greedy Online Gradient Descent (OGD, for $k{=}1$) and lazy GD/dual-averaging (for $k{=}T$), creating a spectrum between reactive…