Imagen de Portada

VisualDeltas: Learning Preferences from Visual Quality Perturbations

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Huang, Hailiang, Liu, Yihao, Guan, Shengyue, Li, Haoze, Li, Sujian
Formato:	Preprint
Publicado:	2026
Materias:	Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

SupChain-Bench: Benchmarking Large Language Models for Real-World Supply Chain Management
por: Guan, Shengyue, et al.
Publicado: (2026)

The Delta Learning Hypothesis: Preference Tuning on Weak Data can Yield Strong Gains
por: Geng, Scott, et al.
Publicado: (2025)

ViPO: Visual Preference Optimization at Scale
por: Li, Ming, et al.
Publicado: (2026)

Revisiting Visual Understanding in Multimodal Reasoning through a Lens of Image Perturbation
por: Li, Yuting, et al.
Publicado: (2025)

Docs2Synth: A Synthetic Data Trained Retriever Framework for Scanned Visually Rich Documents Understanding
por: Ding, Yihao, et al.
Publicado: (2026)

VRD-IU: Lessons from Visually Rich Document Intelligence and Understanding
por: Ding, Yihao, et al.
Publicado: (2025)

Visual Preference Optimization with Rubric Rewards
por: Yu, Ya-Qi, et al.
Publicado: (2026)

When Should We Prefer State-to-Visual DAgger Over Visual Reinforcement Learning?
por: Mu, Tongzhou, et al.
Publicado: (2024)

Thinking with Deltas: Incentivizing Reinforcement Learning via Differential Visual Reasoning Policy
por: Gao, Shujian, et al.
Publicado: (2026)

Adversarial Training with OCR Modality Perturbation for Scene-Text Visual Question Answering
por: Shen, Zhixuan, et al.
Publicado: (2024)

VERTIGO: Visual Preference Optimization for Cinematic Camera Trajectory Generation
por: Li, Mengtian, et al.
Publicado: (2026)

Mitigating Visual Hallucinations via Semantic Curriculum Preference Optimization in MLLMs
por: Li, Yuanshuai, et al.
Publicado: (2025)

Decomposing the Delta: What Do Models Actually Learn from Preference Pairs?
por: Lee, Chia-Hsuan, et al.
Publicado: (2026)

EERPD: Leveraging Emotion and Emotion Regulation for Improving Personality Detection
por: Li, Zheng, et al.
Publicado: (2024)

Shapley Value-based Contrastive Alignment for Multimodal Information Extraction
por: Luo, Wen, et al.
Publicado: (2024)

What External Knowledge is Preferred by LLMs? Characterizing and Exploring Chain of Evidence in Imperfect Context for Multi-Hop QA
por: Chang, Zhiyuan, et al.
Publicado: (2024)

Hidden in Plain Sight: Visual-to-Symbolic Analytical Solution Inference from Field Visualizations
por: Li, Pengze, et al.
Publicado: (2026)

Evaluating LLM-based Agents for Multi-Turn Conversations: A Survey
por: Guan, Shengyue, et al.
Publicado: (2025)

StyleGuard: Preventing Text-to-Image-Model-based Style Mimicry Attacks by Style Perturbations
por: Li, Yanjie, et al.
Publicado: (2025)

The Good, The Bad, and The Greedy: Evaluation of LLMs Should Not Ignore Non-Determinism
por: Song, Yifan, et al.
Publicado: (2024)

Visual Instruction Bottleneck Tuning
por: Oh, Changdae, et al.
Publicado: (2025)

KUDA: Keypoints to Unify Dynamics Learning and Visual Prompting for Open-Vocabulary Robotic Manipulation
por: Liu, Zixian, et al.
Publicado: (2025)

PerturboLLaVA: Reducing Multimodal Hallucinations with Perturbative Visual Training
por: Chen, Cong, et al.
Publicado: (2025)

CloDS: Visual-Only Unsupervised Cloth Dynamics Learning in Unknown Conditions
por: Zhan, Yuliang, et al.
Publicado: (2026)

Parallel Reinforcement Learning Simulation for Visual Quadrotor Navigation
por: Saunders, Jack, et al.
Publicado: (2022)

AI-generated Image Quality Assessment in Visual Communication
por: Tian, Yu, et al.
Publicado: (2024)

RPO: Fine-Tuning Visual Generative Models via Rich Vision-Language Preferences
por: Zhao, Hanyang, et al.
Publicado: (2025)

VQA$^2$: Visual Question Answering for Video Quality Assessment
por: Jia, Ziheng, et al.
Publicado: (2024)

Investigating Anisotropy in Visual Grounding under Controlled Counterfactual Perturbations
por: Lombardo, Gabriele, et al.
Publicado: (2026)

BitAbuse: A Dataset of Visually Perturbed Texts for Defending Phishing Attacks
por: Lee, Hanyong, et al.
Publicado: (2025)

Adaptive Masking Enhances Visual Grounding
por: Jia, Sen, et al.
Publicado: (2024)

Reinforcement Learning for Chain of Thought Compression with One-Domain-to-All Generalization
por: Li, Hanyu, et al.
Publicado: (2025)

Harmful Visual Content Manipulation Matters in Misinformation Detection Under Multimedia Scenarios
por: Wang, Bing, et al.
Publicado: (2026)

Preferred-Action-Optimized Diffusion Policies for Offline Reinforcement Learning
por: Zhang, Tianle, et al.
Publicado: (2024)

Retrieval-Augmented Fine-Tuning With Preference Optimization For Visual Program Generation
por: Kang, Deokhyung, et al.
Publicado: (2025)

Back to Parsimonious Latents: Learning Task-Centric World Models from Visual Foundations
por: Fu, Minghao, et al.
Publicado: (2026)

Cerberus: Efficient Inference with Adaptive Parallel Decoding and Sequential Knowledge Enhancement
por: Liu, Yuxuan, et al.
Publicado: (2024)

Hit-RAG: Learning to Reason with Long Contexts via Preference Alignment
por: Liu, Junming, et al.
Publicado: (2026)

Extended Inductive Reasoning for Personalized Preference Inference from Behavioral Signals
por: Li, Jia-Nan, et al.
Publicado: (2025)

Revisiting Plasticity in Visual Reinforcement Learning: Data, Modules and Training Stages
por: Ma, Guozheng, et al.
Publicado: (2023)