arXiv:2302.05289

Rumor Classification through a Multimodal Fusion Framework and Ensemble Learning

Abderrazek Azri, Cécile Favre, Nouria Harbi, Jérôme Darmont, Camille Noûs

제출 2023-01-04

원문 초록

The proliferation of rumors on social media has become a major concern due to its ability to create a devastating impact. Manually assessing the veracity of social media messages is a very time-consuming task that can be much helped by machine learning. Most message veracity verification methods only exploit textual contents and metadata. Very few take both textual and visual contents, and more particularly images, into account. Moreover, prior works have used many classical machine learning models to detect rumors. However, although recent studies have proven the effectiveness of ensemble machine learning approaches, such models have seldom been applied. Thus, in this paper, we propose a set of advanced image features that are inspired from the field of image quality assessment, and introduce the Multimodal fusiON framework to assess message veracIty in social neTwORks (MONITOR), which exploits all message features by exploring various machine learning models. Moreover, we demonstrate the effectiveness of ensemble learning algorithms for rumor detection by using five metalearning models. Eventually, we conduct extensive experiments on two real-world datasets. Results show that MONITOR outperforms state-of-the-art machine learning baselines and that ensemble models significantly increase MONITOR's performance.

함께 살펴볼 논문

공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.

arXiv:2607.00416v1cs.CV공통 태그 Ensemble Learning, Image Quality Assessment

DroneIQA-VLE: 비전-언어 앙상블을 통한 멀티태스크 드론 이미지 화질 평가

DroneIQA-VLE: Multi-Task Drone Image Quality Assessment via Vision-Language Ensemble

저고도 UAV 이미지에 대한 타겟 인식 화질 평가를 주제로 한 ICME 2026 Drone-IQA Grand Challenge의 솔루션인 DroneIQA-VLE를 제안한다. 이 프레임워크는 두 개의 상호 보완적인 파이프라인을 앙상블하여 전역(global), 타겟(target), 배경(background) 화질 점수를 공동으로 예측한다: (1) 멀티태스크 회귀 헤드를 갖춘 SigLIP2 비전 인코더, (2) 화질 점수 회귀를 위해 LoRA로 적응된 Qwen3.5-9B 멀티모달 대규모 언어 모델(LLM). 최종 전역 화질 예측은 두 파이프라인의 출력을 산술 평균하여 얻는다. 제안 방법은 챌린지에서 2위를 달성하여 그 효과를 입증하였다. 코드는 https://github.com/sunwei925/DroneIQA-VLE 에서 확인할 수 있다.

토론

이 논문과 연결된 포럼 글이 아직 없습니다.