Seeing Far and Clearly: Mitigating Hallucinations in MLLMs with Attention Causal Decoding

Recent advancements in multimodal large language models (MLLMs) have significantly improved performance in visual question answering. However, they often suffer from hallucinations. In this work, hallucinations are categorized into two main types: initial hallucinations and snowball hallucinations....

Full description

Saved in:
Bibliographic Details
Published in:Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) Vol. 2025; pp. 26147 - 26159
Main Authors: Tang, Feilong, Liu, Chengzhi, Xu, Zhongxing, Hu, Ming, Huang, Zile, Xue, Haochen, Chen, Ziyang, Peng, Zelin, Yang, Zhiwei, Zhou, Sijin, Li, Wenxue, Li, Yulong, Song, Wenxuan, Su, Shiyan, Feng, Wei, Su, Jionglong, Lin, Minquan, Peng, Yifan, Cheng, Xuelian, Razzak, Imran, Ge, Zongyuan
Format: Conference Proceeding Journal Article
Language:English
Published: United States IEEE 01.06.2025
Subjects:
Data mining
Decoding
Encoding
FarSight demonstrates significant hallucination-mitigating performance across different MLLMs on both image and video benchmarks
Heart
Interference
Large language models
proving its effectiveness
Question answering (information retrieval)
Registers
Video sequences
Visualization
With extensive experiments
ISSN:1063-6919, 1063-6919
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Recent advancements in multimodal large language models (MLLMs) have significantly improved performance in visual question answering. However, they often suffer from hallucinations. In this work, hallucinations are categorized into two main types: initial hallucinations and snowball hallucinations. We argue that adequate contextual information can be extracted directly from the token interaction process. Inspired by causal inference in the decoding strategy, we propose to leverage causal masks to establish information propagation between multimodal tokens. The hypothesis is that insufficient interaction between those tokens may lead the model to rely on outlier tokens, overlooking dense and rich contextual cues. Therefore, we propose to intervene in the propagation process by tackling outlier tokens to enhance in-context inference. With this goal, we present FarSight, a versatile plug-and-play decoding strategy to reduce attention interference from outlier tokens merely by optimizing the causal mask. The heart of our method is effective token propagation. We design an attention register structure within the upper triangular matrix of the causal mask, dynamically allocating attention to capture attention diverted to outlier tokens. Moreover, a positional awareness encoding method with a diminishing masking rate is proposed, allowing the model to attend to further preceding tokens, especially for video sequence tasks.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1063-6919
1063-6919
DOI:10.1109/CVPR52734.2025.02435