paper-with-me

홈 › Papers

Efficient Contrastive Decoding with Probabilistic Hallucination Detection - Mitigating Hallucinations in Large Vision Language Models -

2025-04-16 · Laura Fieback, Nishilkumar Balar, Jakob Spiegelberg, Hanno Gottschalk

Despite recent advances in Large Vision Language Models (LVLMs), these models still suffer from generating hallucinatory responses that do not align with the visual input provided. To mitigate such hallucinations, we introduce Efficient Contrastive Decoding (ECD), a simple method that leverages probabilistic hallucination detection to shift the output distribution towards contextually accurate answers at inference time. By contrasting token probabilities and hallucination scores, ECD subtracts hallucinated concepts from the original distribution, effectively suppressing hallucinations. Notably, our proposed method can be applied to any open-source LVLM and does not require additional LVLM training. We evaluate our method on several benchmark datasets and across different LVLMs. Our experiments show that ECD effectively mitigates hallucinations, outperforming state-of-the-art methods with respect to performance on LVLM benchmarks and computation time.

📄 PDF Abstract BibTeX arXiv:2504.12137

Code (0)

등록된 구현이 없습니다.

Tasks

Hallucination

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

The Mirage of Performance Gains: Why Contrastive Decoding Fails to Address Multimodal Hallucination

2025-04-14 · Hao Yin, Guangzong Si, Zilei Wang

Contrastive decoding strategies are widely used to reduce hallucinations in multimodal large language models (MLLMs). These methods work by constructing contrastive samples to induce hallucinations and then suppressing t…

Hallucination

Rethinking CD: A Reproducibility Study and Extension on the Ineffectiveness of Contrastive Decoding at Mitigating Object Hallucinations in MLLMs

2026-07-28 · Arnav Bendre, Guneesh Gupta, Kavish Grover, Chayan Aggarwal 외 arxiv

Contrastive decoding (CD) has been proposed as a training-free strategy for mitigating object hallucinations in multimodal large language models (MLLMs), with reported gains on benchmarks such as POPE. However, recent wo…

Visual Grounding

ASCD: Attention-Steerable Contrastive Decoding for Reducing Hallucination in MLLM

2025-06-17 · Yujun Wang, Jinhe Bi, Yunpu Ma, Soeren Pirk

Multimodal Large Language Model (MLLM) often suffer from hallucinations. They over-rely on partial cues and generate incorrect responses. Recently, methods like Visual Contrastive Decoding (VCD) and Instruction Contrasti…

HallucinationLanguage ModelingLanguage ModellingLarge Language Model+2

ConVis: Contrastive Decoding with Hallucination Visualization for Mitigating Hallucinations in Multimodal Large Language Models

2024-08-25 · Yeji Park, Deokyeong Lee, Junsuk Choe, Buru Chang

Hallucinations in Multimodal Large Language Models (MLLMs) where generated responses fail to accurately reflect the given image pose a significant challenge to their reliability. To address this, we introduce ConVis, a n…

Hallucination

MaskCD: Mitigating LVLM Hallucinations by Image Head Masked Contrastive Decoding

2025-10-03 · Jingyuan Deng, Yujiu Yang arxiv

Large vision-language models (LVLMs) have shown remarkable performance in visual-language understanding for downstream multimodal tasks. While their capabilities are improving, problems emerge simultaneously. Among those…