Uncovering Bias in Large Vision-Language Models with Counterfactuals
- URL: http://arxiv.org/abs/2404.00166v2
- Date: Fri, 7 Jun 2024 23:29:19 GMT
- Title: Uncovering Bias in Large Vision-Language Models with Counterfactuals
- Authors: Phillip Howard, Anahita Bhiwandiwalla, Kathleen C. Fraser, Svetlana Kiritchenko,
- Abstract summary: We study the social biases contained in text generated by Large Vision-Language Models (LVLMs)
We present LVLMs with identical open-ended text prompts while conditioning on images from different counterfactual sets.
We find that social attributes such as race, gender, and physical characteristics depicted in input images can significantly influence toxicity and the generation of competency-associated words.
- Score: 8.414108895243148
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: With the advent of Large Language Models (LLMs) possessing increasingly impressive capabilities, a number of Large Vision-Language Models (LVLMs) have been proposed to augment LLMs with visual inputs. Such models condition generated text on both an input image and a text prompt, enabling a variety of use cases such as visual question answering and multimodal chat. While prior studies have examined the social biases contained in text generated by LLMs, this topic has been relatively unexplored in LVLMs. Examining social biases in LVLMs is particularly challenging due to the confounding contributions of bias induced by information contained across the text and visual modalities. To address this challenging problem, we conduct a large-scale study of text generated by different LVLMs under counterfactual changes to input images. Specifically, we present LVLMs with identical open-ended text prompts while conditioning on images from different counterfactual sets, where each set contains images which are largely identical in their depiction of a common subject (e.g., a doctor), but vary only in terms of intersectional social attributes (e.g., race and gender). We comprehensively evaluate the text produced by different LVLMs under this counterfactual generation setting and find that social attributes such as race, gender, and physical characteristics depicted in input images can significantly influence toxicity and the generation of competency-associated words.
Related papers
- SEED-Story: Multimodal Long Story Generation with Large Language Model [66.37077224696242]
SEED-Story is a novel method that leverages a Multimodal Large Language Model (MLLM) to generate extended multimodal stories.
We propose multimodal attention sink mechanism to enable the generation of stories with up to 25 sequences (only 10 for training) in a highly efficient autoregressive manner.
We present a large-scale and high-resolution dataset named StoryStream for training our model and quantitatively evaluating the task of multimodal story generation in various aspects.
arXiv Detail & Related papers (2024-07-11T17:21:03Z) - Rethinking Visual Prompting for Multimodal Large Language Models with External Knowledge [76.45868419402265]
multimodal large language models (MLLMs) have made significant strides by training on vast high-quality image-text datasets.
However, the inherent difficulty in explicitly conveying fine-grained or spatially dense information in text, such as masks, poses a challenge for MLLMs.
This paper proposes a new visual prompt approach to integrate fine-grained external knowledge, gleaned from specialized vision models, into MLLMs.
arXiv Detail & Related papers (2024-07-05T17:43:30Z) - Evaluating Fairness in Large Vision-Language Models Across Diverse Demographic Attributes and Prompts [27.66626125248612]
We empirically investigate emphvisual fairness in several mainstream vision-language models (LVLMs)
We audit their performance disparities across sensitive demographic attributes based on public fairness benchmark datasets (e.g., FACET)
Despite enhancements in visual understanding, both open-source and closed-source LVLMs exhibit prevalent fairness issues across different instruct prompts and demographic attributes.
arXiv Detail & Related papers (2024-06-25T23:11:39Z) - Uncovering Bias in Large Vision-Language Models at Scale with Counterfactuals [8.41410889524315]
We study the social biases contained in text generated by Large Vision-Language Models (LVLMs)
We present LVLMs with identical open-ended text prompts while conditioning on images from different counterfactual sets.
We evaluate the text produced by different models under this counterfactual generation setting at scale, producing over 57 million responses from popular LVLMs.
arXiv Detail & Related papers (2024-05-30T15:27:56Z) - InternLM-XComposer2: Mastering Free-form Text-Image Composition and
Comprehension in Vision-Language Large Model [108.42241250772643]
We introduce InternLM-XComposer2, a vision-language model excelling in free-form text-image composition and comprehension.
This model goes beyond conventional vision-language understanding, adeptly crafting interleaved text-image content from diverse inputs.
Experimental results demonstrate the superiority of InternLM-XComposer2 based on InternLM2-7B in producing high-quality long-text multi-modal content.
arXiv Detail & Related papers (2024-01-29T18:59:02Z) - StoryGPT-V: Large Language Models as Consistent Story Visualizers [39.790319429455856]
generative models have demonstrated impressive capabilities in generating realistic and visually pleasing images grounded on textual prompts.
Yet, the emerging Large Language Model (LLM) showcases robust reasoning abilities to navigate through ambiguous references.
We introduce textbfStoryGPT-V, which leverages the merits of the latent diffusion (LDM) and LLM to produce images with consistent and high-quality characters.
arXiv Detail & Related papers (2023-12-04T18:14:29Z) - Unified Language-Vision Pretraining in LLM with Dynamic Discrete Visual Tokenization [52.935150075484074]
We introduce a well-designed visual tokenizer to translate the non-linguistic image into a sequence of discrete tokens like a foreign language.
The resulting visual tokens encompass high-level semantics worthy of a word and also support dynamic sequence length varying from the image.
This unification empowers LaVIT to serve as an impressive generalist interface to understand and generate multi-modal content simultaneously.
arXiv Detail & Related papers (2023-09-09T03:01:38Z) - LLM-CXR: Instruction-Finetuned LLM for CXR Image Understanding and Generation [51.08810811457617]
vision-language alignment in LLMs is actively being researched to enable multimodal reasoning and visual IO.
We develop a method for instruction-tuning an LLM only on text to gain vision-language capabilities for medical images.
Our model, LLM-CXR, trained in this approach shows better image-text alignment in both CXR understanding and generation tasks.
arXiv Detail & Related papers (2023-05-19T07:44:39Z) - Visually-Augmented Language Modeling [137.36789885105642]
We propose a novel pre-training framework, named VaLM, to Visually-augment text tokens with retrieved relevant images for Language Modeling.
With the visually-augmented context, VaLM uses a visual knowledge fusion layer to enable multimodal grounded language modeling.
We evaluate the proposed model on various multimodal commonsense reasoning tasks, which require visual information to excel.
arXiv Detail & Related papers (2022-05-20T13:41:12Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.