paper-with-me

홈 › Papers

DeepSeek on a Trip: Inducing Targeted Visual Hallucinations via Representation Vulnerabilities

2025-02-11 · Chashi Mahiul Islam, Samuel Jacob Chacko, Preston Horne, Xiuwen Liu

Multimodal Large Language Models (MLLMs) represent the cutting edge of AI technology, with DeepSeek models emerging as a leading open-source alternative offering competitive performance to closed-source systems. While these models demonstrate remarkable capabilities, their vision-language integration mechanisms introduce specific vulnerabilities. We implement an adapted embedding manipulation attack on DeepSeek Janus that induces targeted visual hallucinations through systematic optimization of image embeddings. Through extensive experimentation across COCO, DALL-E 3, and SVIT datasets, we achieve hallucination rates of up to 98.0% while maintaining high visual fidelity (SSIM > 0.88) of the manipulated images on open-ended questions. Our analysis demonstrates that both 1B and 7B variants of DeepSeek Janus are susceptible to these attacks, with closed-form evaluation showing consistently higher hallucination rates compared to open-ended questioning. We introduce a novel multi-prompt hallucination detection framework using LLaMA-3.1 8B Instruct for robust evaluation. The implications of these findings are particularly concerning given DeepSeek's open-source nature and widespread deployment potential. This research emphasizes the critical need for embedding-level security measures in MLLM deployment pipelines and contributes to the broader discussion of responsible AI implementation.

📄 PDF Abstract BibTeX arXiv:2502.07905

Code (0)

등록된 구현이 없습니다.

Tasks

HallucinationSSIM

Similar Papers 제목 키워드 기반

HICD: Hallucination-Inducing via Attention Dispersion for Contrastive Decoding to Mitigate Hallucinations in Large Language Models

2025-03-17 · Xinyan Jiang, Hang Ye, Yongxin Zhu, Xiaoying Zheng 외

Large Language Models (LLMs) often generate hallucinations, producing outputs that are contextually inaccurate or factually incorrect. We introduce HICD, a novel method designed to induce hallucinations for contrastive d…

HallucinationQuestion AnsweringReading Comprehension

Alleviating Hallucinations in Large Vision-Language Models through Hallucination-Induced Optimization

2024-05-24 · Xinyu Lyu, Beitao Chen, Lianli Gao, Jingkuan Song 외

Although Large Visual Language Models (LVLMs) have demonstrated exceptional abilities in understanding multimodal data, they invariably suffer from hallucinations, leading to a disconnect between the generated text and t…

Hallucination

Gradient-guided Attention Map Editing: Towards Efficient Contextual Hallucination Mitigation

2025-03-11 · Yu Wang, Jiaxin Zhang, Xiang Gao, Wendi Cui 외

In tasks like summarization and open-book question answering (QA), Large Language Models (LLMs) often encounter "contextual hallucination", where they produce irrelevant or incorrect responses despite having access to ac…

Computational EfficiencyHallucinationQuestion Answering

HII-DPO: Eliminate Hallucination via Accurate Hallucination-Inducing Counterfactual Images

2026-02-11 · Yilin Yang, Zhenghui Guo, Yuke Wang, Omprakash Gnawali 외 arxiv

Large Vision-Language Models (VLMs) have achieved remarkable success across diverse multimodal tasks but remain vulnerable to hallucinations rooted in inherent language bias. Despite recent progress, existing hallucinati…

Mitigating Hallucination in Multimodal Large Language Model via Hallucination-targeted Direct Preference Optimization

2024-11-15 · Yuhan Fu, Ruobing Xie, Xingwu Sun, Zhanhui Kang 외

Multimodal Large Language Models (MLLMs) are known to hallucinate, which limits their practical applications. Recent works have attempted to apply Direct Preference Optimization (DPO) to enhance the performance of MLLMs,…

HallucinationHallucination EvaluationLanguage ModelingLanguage Modelling+2