paper-with-me

Papers

CATCH: Complementary Adaptive Token-level Contrastive Decoding to Mitigate Hallucinations in LVLMs

2024-11-19 · Zhehan Kan, Ce Zhang, Zihan Liao, Yapeng Tian, Wenming Yang, Junyuan Xiao, Xu Li, Dongmei Jiang, YaoWei Wang, Qingmin Liao

Large Vision-Language Model (LVLM) systems have demonstrated impressive vision-language reasoning capabilities but suffer from pervasive and severe hallucination issues, posing significant risks in critical domains such as healthcare and autonomous systems. Despite previous efforts to mitigate hallucinations, a persistent issue remains: visual defect from vision-language misalignment, creating a bottleneck in visual processing capacity. To address this challenge, we develop Complementary Adaptive Token-level Contrastive Decoding to Mitigate Hallucinations in LVLMs (CATCH), based on the Information Bottleneck theory. CATCH introduces Complementary Visual Decoupling (CVD) for visual information separation, Non-Visual Screening (NVS) for hallucination detection, and Adaptive Token-level Contrastive Decoding (ATCD) for hallucination mitigation. CATCH addresses issues related to visual defects that cause diminished fine-grained feature perception and cumulative hallucinations in open-ended scenarios. It is applicable to various visual question-answering tasks without requiring any specific data or prior knowledge, and generalizes robustly to new tasks without additional training, opening new possibilities for advancing LVLM in various challenging applications.

📄 PDF Abstract BibTeX arXiv:2411.12713

Code (0)

등록된 구현이 없습니다.

Tasks

HallucinationLanguage ModelingLanguage ModellingQuestion AnsweringVisual Question Answering

Similar Papers 제목 키워드 기반

Bi-Granularity Contrastive Learning for Post-Training in Few-Shot Scene

2021-06-04 · Findings (ACL) 2021 8 · Ruikun Luo, Guanhuan Huang, Xiaojun Quan

The major paradigm of applying a pre-trained language model to downstream tasks is to fine-tune it on labeled task data, which often suffers instability and low performance when the labeled examples are scarce.~One way t…

Contrastive LearningData AugmentationLanguage ModelingLanguage Modelling+1

Evading Visual Aphasia: Contrastive Adaptive Semantic Token Pruning for Vision-Language Models

2026-05-10 · Jie Ma, Yihang Liu, Zhike Qiu, Jiayi Ji 외 arxiv

Are low-attention visual tokens truly redundant in vision-language reasoning? Existing pruning methods often assume so, ranking visual tokens by shallow text-to-image attention and discarding low-scoring patches to accel…

Visual Grounding

CoME-VL: Scaling Complementary Multi-Encoder Vision-Language Learning

2026-04-03 · Ankan Deria, Komal Kumar, Xilin He, Imran Razzak 외 arxiv

Recent vision-language models (VLMs) typically rely on a single vision encoder trained with contrastive image-text objectives, such as CLIP-style pretraining. While contrastive encoders are effective for cross-modal alig…

ASPECT: Node-Level Adaptive Spectral Fusion for Graph Contrastive Learning

2026-04-02 · Zhuolong Li, Boxue Yang, Haopeng Chen arxiv

Spectral graph contrastive learning often constructs low- and high-frequency views to capture complementary graph signals, but these views are commonly combined by graph-level or node-agnostic fusion rules. We show that …

Contrastive Learning

Please refuse to answer me! Mitigating Over-Refusal in Large Language Models via Adaptive Contrastive Decoding

2026-04-18 · Yupeng Qi, Ziyu Lyu, Lixin Cui, Lu Bai 외 arxiv

Safety-aligned large language models (LLMs) often generate refusal responses to harmless queries due to the over-refusal problem. However, existing methods for mitigating over-refusal cannot maintain a low refusal ratio …