AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

MMBERT: Scaled Mixture-of-Experts Multimodal BERT for Robust Chinese Hate Speech Detection Under Cloaking Perturbations

Qiyao Xue, Yuchen Dou, Zheyuan Ryan Shi, Xiang Lorraine Li, Wei Gao

AAAI 2026 · Vision-language and multimodal › Cross-modal learning · approach: Multimodal sentiment analysis · 7 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add speech modality to text-only MMBERT (ablation)MMBERT · ToxiCloakCNmacro F1: 86.1 -> 91.1Improved
add vision modality to text-only MMBERT (ablation)MMBERT · ToxiCloakCNmacro F1: 86.1 -> 86.6No clear effect
add speech modality to text-only MMBERT (ablation)MMBERT · COLDmacro F1: 81.5 -> 83.8Improved
add both speech and vision with MoE and 3-stage training (full MMBERT vs text-speech) (ablation)MMBERT · ToxiCloakCNmacro F1: 91.1 -> 95.2Improved
add progressive three-stage training (aligner and expert stages) (ablation)MMBERT · ToxiCloakCNvalidation accuracy / training lossImproved
use MMBERT instead of ChineseBERTfine-tuned encoders · ToxiCloakCNmacro F1: 86.8 -> 95.2Improved
use MMBERT instead of ChineseBERTfine-tuned encoders · ToxiCNmacro F1: 90.6 -> 92.2Improved
add perturbed examples and CoT to GPT-4o promptGPT-4o · ToxiCloakCNmacro F1: 62.4 -> 69.3Improved
use CoT prompting and perturbed examplesDeepSeek-v3 · ToxiCNmacro F1: 71.7 -> 77.1 (basic to CoT) while ToxiCloakCN 66.2 -> 71.6No clear effect
fine-tune LLaMA3-8B vs prompting LLaMA3-8B (2 examples)LLaMA3-8B · ToxiCloakCNmacro F1: 68.0 -> 79.3Improved

Later papers that compared with this method

1 later accepted paper reports MMBERT as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy