Image de couverture de livre

Dissecting and Mitigating Diffusion Bias via Mechanistic Interpretability

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Shi, Yingdong, Li, Changming, Wang, Yifan, Zhao, Yongxiang, Pang, Anqi, Yang, Sibei, Yu, Jingyi, Ren, Kan
Format:	Preprint
Publié:	2025
Sujets:	Computer Vision and Pattern Recognition Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Discovering Influential Neuron Path in Vision Transformers
par: Wang, Yifan, et autres
Publié: (2025)

Mitigating Bias in Concept Bottleneck Models for Fair and Interpretable Image Classification
par: Tong, Schrasing, et autres
Publié: (2026)

The devil is in the object boundary: towards annotation-free instance segmentation using Foundation Models
par: Shi, Cheng, et autres
Publié: (2024)

InvDiff: Invariant Guidance for Bias Mitigation in Diffusion Models
par: Hou, Min, et autres
Publié: (2024)

Vision Function Layer in Multimodal LLMs
par: Shi, Cheng, et autres
Publié: (2025)

Vision Transformers Need More Than Registers
par: Shi, Cheng, et autres
Publié: (2026)

Penalizing Boundary Activation for Object Completeness in Diffusion Models
par: Xu, Haoyang, et autres
Publié: (2025)

PriSampler: Mitigating Property Inference of Diffusion Models
par: Hu, Hailong, et autres
Publié: (2023)

SeqAfford: Sequential 3D Affordance Reasoning via Multimodal Large Language Model
par: Yu, Chunlin, et autres
Publié: (2024)

Context-Aware Aerial Object Detection: Leveraging Inter-Object and Background Relationships
par: Ren, Botao, et autres
Publié: (2024)

Frequency Regulation for Exposure Bias Mitigation in Diffusion Models
par: Yu, Meng, et autres
Publié: (2025)

OMG: Towards Open-vocabulary Motion Generation via Mixture of Controllers
par: Liang, Han, et autres
Publié: (2023)

Chart Deep Research in LVLMs via Parallel Relative Policy Optimization
par: Tang, Jiajin, et autres
Publié: (2026)

Dissecting Representation Misalignment in Contrastive Learning via Influence Function
par: Hu, Lijie, et autres
Publié: (2024)

Debiasify: Self-Distillation for Unsupervised Bias Mitigation
par: Bayasi, Nourhan, et autres
Publié: (2024)

WeaveTime: Stream from Earlier Frames into Emergent Memory in VideoLLMs
par: Zhang, Yulin, et autres
Publié: (2026)

Plain-Det: A Plain Multi-Dataset Object Detector
par: Shi, Cheng, et autres
Publié: (2024)

A novel spatial-frequency domain network for zero-shot incremental learning
par: Ren, Jie, et autres
Publié: (2024)

Controllable Feature Whitening for Hyperparameter-Free Bias Mitigation
par: Cho, Yooshin, et autres
Publié: (2025)

Mitigating the Bias in the Model for Continual Test-Time Adaptation
par: Chung, Inseop, et autres
Publié: (2024)

GPO-V: Jailbreak Diffusion Vision Language Model by Global Probability Optimization
par: Pan, Yu, et autres
Publié: (2026)

Learning Single Index Models with Diffusion Priors
par: Tang, Anqi, et autres
Publié: (2025)

Mitigating Bias in Dataset Distillation
par: Cui, Justin, et autres
Publié: (2024)

A Unified and Fast-Sampling Diffusion Bridge Framework via Stochastic Optimal Control
par: Pan, Mokai, et autres
Publié: (2025)

Mitigating Long-Tail Bias via Prompt-Controlled Diffusion Augmentation
par: Wijenayake, Buddhi, et autres
Publié: (2026)

BM-CL: Bias Mitigation through the lens of Continual Learning
par: Mansilla, Lucas, et autres
Publié: (2025)

Curriculum Point Prompting for Weakly-Supervised Referring Image Segmentation
par: Dai, Qiyuan, et autres
Publié: (2024)

Free on the Fly: Enhancing Flexibility in Test-Time Adaptation with Online EM
par: Dai, Qiyuan, et autres
Publié: (2025)

FEB-Cache: Frequency-Guided Exposure Bias Reduction for Enhancing Diffusion Transformer Caching
par: Zou, Zhen, et autres
Publié: (2025)

LILAC: Long-sequence Incremental Low-latency Arbitrary Motion Stylization via Streaming VAE-Diffusion with Causal Decoding
par: Ren, Peng, et autres
Publié: (2025)

Identify, Isolate, and Purge: Mitigating Hallucinations in LVLMs via Self-Evolving Distillation
par: Li, Wenhao, et autres
Publié: (2025)

Mechanistic Interpretability for Learning Assurance of a Vision-Based Landing System
par: Valentin, Romeo, et autres
Publié: (2026)

Eyes Wide Open: Ego Proactive Video-LLM for Streaming Video
par: Zhang, Yulin, et autres
Publié: (2025)

Interpreting and Editing Vision-Language Representations to Mitigate Hallucinations
par: Jiang, Nick, et autres
Publié: (2024)

Semi-Supervised Transfer Boosting (SS-TrBoosting)
par: Deng, Lingfei, et autres
Publié: (2024)

How I Met Your Bias: Investigating Bias Amplification in Diffusion Models
par: Roos, Nathan, et autres
Publié: (2025)

Mitigating Algorithmic Bias in Multiclass CNN Classifications Using Causal Modeling
par: Byun, Min Sik, et autres
Publié: (2025)

SWiFT: Soft-Mask Weight Fine-tuning for Bias Mitigation
par: Yan, Junyu, et autres
Publié: (2025)

Mitigating Spurious Background Bias in Multimedia Recognition with Disentangled Concept Bottlenecks
par: Huang, Gaoxiang, et autres
Publié: (2025)

Intervene-All-Paths: Unified Mitigation of LVLM Hallucinations across Alignment Formats
par: Qian, Jiaye, et autres
Publié: (2025)