Curated Data for Efficient Learning

Archival Track

VolDoGer: LLM-assisted Datasets for Domain Generalization in Vision-Language Tasks

Authors: Juhwan Choi (AITRICS); Junehyoung Kwon (Chung-Ang University); JungMin Yun (Chung-Ang University); Seunguk Yu (Chung-Ang University); YoungBin Kim (Chung-Ang University)
arXiv

Objaverse++: Curated 3D Object Dataset with Quality Annotations

Authors: Chendi Lin (CMU); Heshan Liu (CMU); Qunshu Lin (Zhejiang University); Zachary Bright (Exascale Labs); Shitao Tang (Simon Fraser University); Yihui He (CMU); Minghao Liu (2077AI); Ling Zhu (Exascale Labs); Cindy Le (Google)
arXiv | Project Page

Few-Shot Vision-Language Reasoning for Satellite Imagery via Verifiable Rewards

Authors: Aybora Köksal (METU); A. Aydın Alatan (METU)
arXiv

LG-Traj: LLM Guided Pedestrian Trajectory Prediction

Authors: Pranav singh chib (IIT Roorkee); Pravendra Singh (IIT Roorkee)
PDF

Federated Active Learning for Target Domain Generalisation

Authors: Razvan Caramalau (UCL); Binod Bhattarai (Aberdeen); Danail Stoyanov (UCL)
arXiv

DISC-GAN: Disentangling Style and Content for Cluster-Specific Synthetic Underwater Image Generation

Authors: Sneha Varur (KLE Technological University); Anirudh R Hanchinamani (KLE Technological University); Tarun S Bagewadi (KLE Technological University); Uma Mudenagudi (KLE Technological University), Chaitra D Desai (KLE Technological University), Sujata C (KLE Technological University); Padmashree Desai (KLE Technological University); Sumit Meharwade (KLE Technological University);
PDF

OmViD: Omni-supervised active learning for video action detection

Authors: Aayush Rana (Qualcomm), Akash Kumar (UCF), Vibhav Vineet (Microsoft), Yogesh S Rawat (UCF)
arXiv | Project Page

Class-Proportional Coreset Selection for Difficulty-Separable Data

Authors: Elisa Tsai (Michigan), Haizhong Zheng (CMU), Atul Prakash (Michigan)
arXiv

Task-Specific Generative Dataset Distillation with Difficulty-Guided Sampling

Authors: Mingzhuo Li (Hokkaido University), Guang Li(Hokkaido University), Jiafeng Mao (The University of Tokyo), Linfeng Ye (University of Toronto), Takahiro Ogawa (Hokkaido University), Miki Haseyama (Hokkaido University)
arXiv | Project Page

Synthetic Captions for Open-Vocabulary Zero-Shot Segmentation

Authors: Tim Lebailly (Meta, KU Leuven); Vijay Veerabadran (Meta); Satwik Kottur (Meta); Karl Ridgeway (Meta); Michael Louis Iuzzolino (Meta)
arXiv

Efficient Learning for Product Attributes with Compact Multimodal Models

Authors: Mandar Kulkarni (Flipkart)
arXiv

SYM3D: Canonicalizing Triplanes via Symmetry for Single-View 3D Learning

Authors: Jing Yang, Kyle Fogarty, Fangcheng Zhong, Cengiz Oztireli (Cambridge)
PDF

How to Train your Text-to-Image Model: Evaluating Design Choices for Synthetic Training Captions

Authors: Manuel Brack, Sudeep Katakol, Felix Friedrich, Patrick Schramowski, Hareesh Ravi, Kristian Kersting, Ajinkya Kale
arXiv

Non-Archival Track

Data-Efficient Learning with Sparse Adversarial Coresets

Authors: Manasa Madabhushi; Tushar Shinde

Adapting Vehicle Detectors for Aerial Imagery to Unseen Domains with Weak Supervision

Authors: Xiao Fang (CMU); Minhyek Jeon (CMU); Zheyang Qin (CMU); Stanislav Panev (CMU); Celso de Melo (DEVCOM); Shuowen Hu (DEVCOM); Shayok Chakraborty (CMU, FSU); Fernando De la Torre (CMU)
arXiv | Project Page

A CLIP-Powered Framework for Robust and Generalizable Data Selection;Dynamic Soft Data Pruning: Adaptive Selection and Scheduling for Data-Efficient Learning

Authors: Suorong Yang (NJU); Junda Yu (FDU); Jucheng Hu (UCL);
arXiv

Making Something from (Almost) Nothing: Extreme Low-Resource Visual Learning with Diffusion Synthesis and Self-Supervised Distillation

Authors: Xuying Li (Hydrox AI)
PDF

ControlTac: Force- and Position-Controlled Tactile Data Augmentation with a Single Reference Image

Authors: Dongyu Luo(UMD, HKU), Kelin Yu(UMD), Amir Hossein Shahidzadeh(UMD), Cornelia Fermuller(UMD), Yiannis Aloimonos(UMD), Ruohan Gao(UMD)
PDF | Project Page

RF PRIOR: Preserving Global-Context Priors for Efficient Instance Segmentation Transfer

Authors: Jason K. Nam (Hongik University, ETRI)

VILA$^2$: Towards VLM Augmentation via Self-Improvement

Authors: Yunhao Fang (Nvidia), Ligeng Zhu (Nvidia), Yao Lu (Nvidia), Yan Wang (Nvidia), Pavlo Molchanov (Nvidia), Jan Kautz (Nvidia), Jang Hyun Cho (UT Austin), Marco Pavone (Nvidia), Song Han (Nvidia, MIT), Hongxu Yin (Nvidia)
arXiv

SIDA: Synthetic Image Driven Zero-shot Domain Adaptation

Authors: Ye-Chan Kim (Hanyang University); SeungJu Cha (Hanyang University); Si-Woo Kim (Hanyang University); Taewhan Kim (Hanyang University); Dong-Jin Kim (Hanyang University)
arXiv

arXivBench: A Curated Benchmark for Evaluating LLM Accuracy in Assisting Academic Writing

Authors: Ning Li (UCLA); Jingran Zhang (UCLA), Justin Cui (UCLA)
arXiv | Project Page

Data-Efficient Ensemble Weather Forecasting with Diffusion Models

Authors: Kevin Valencia (UCLA), Ziyang Liu (UCLA), Justin Cui (UCLA)
arXiv | PDF

Generating Fine Details of Entity Interactions

Authors: Xinyi Gu (MIT); Jiayuan Mao (MIT)
arXiv | Project Page

Dynamic Soft Data Pruning: Adaptive Selection and Scheduling for Data-Efficient Learning

Authors: Junda Yu (FDU); Suorong Yang (NJU); Jucheng Hu (UCL); Dongzhan Zhou (SHLab); Peng Ye (FDU); Tao Chen (FDU)

SynC: Synthetic Image Caption Dataset Refinement with One-to-many Mapping for Zero-shot Image Captioning

Authors: Si-Woo Kim (Hanyang University); MinJu Jeon (Hanyang University); Ye-Chan Kim (Hanyang University); Soeun Lee(AI R&D Division, CJ Group); Taewhan Kim (Hanyang University); Dong-Jin Kim (Hanyang University)
arXiv

Role Bias in Text-to-Image Diffusion Models: Diagnosing and Mitigating Compositional Failures through Intermediate Decomposition

Authors: Sina Malakouti (University of Pittsburgh); Adriana Kovashka (University of Pittsburgh)
arXiv

AnyBald: Toward Realistic Diffusion-Based Hair Removal In-The-Wild

Authors: Yongjun Choi(UNIST), Seungoh Han(UNIST), Soomin Kim(Ewha Woman’s University), Sumin Son(Ewha Woman’s University), Mohsen Rohani(L’Oréal), Edgar Maucourant(L’Oréal), Dongbo Min(Ewha Woman’s University), Kyungdon Joo(UNIST)
PDF

Efficient Long-Tail Learning in Latent Space by sampling Synthetic Data

Authors: Nakul Sharma (Independent Researcher)
PDF

Autoguided Online Data Curation for Diffusion Model Training

Authors: Valeria Pais (University of Glasgow); Luis Oala (Dotphoton); Daniele Faccio (University of Glasgow); Marco Aversa (Dotphoton)
PDF

Learning Hyperspectral Images with Curated Text Prompts for Efficient Multimodal Alignment

Authors: Abhiroop Chatterjee (Jadavpur University); Susmita Ghosh (Jadavpur University)
PDF

ChartGen: Scaling Chart Understanding Via Code-Guided Synthetic Chart Generation

Authors: Jovana Kondic (MIT), Pengyuan Li (IBM Research), Dhiraj Joshi (IBM Research), Zexue He (MIT-IBM Watson AI Labs), Shafiq Abedin (IBM Research), Jennifer Sun (MIT), Ben Wiesel (IBM Research), Eli Schwartz (IBM Research), Ahmed Nassar (IBM Research), Bo Wu (MIT-IBM Watson AI Labs, IBM Research), Assaf Arbelle (IBM Research), Aude Oliva (MIT, MIT-IBM Watson AI Labs), Dan Gutfreund (MIT-IBM Watson AI Labs, IBM Research), Leonid Karlinsky (MIT-IBM Watson AI Labs, IBM Research), Rogerio Feris (MIT-IBM Watson AI Labs, IBM Research)
arXiv

VILA^2: Towards VLM Augmentation via Self-Improvement

Authors: Yunhao Fang (NVIDIA), Ligeng Zhu (NVIDIA), Yao Lu (NVIDIA), Yan Wang (NVIDIA), Pavlo Molchanov (NVIDIA), Jang Hyun Cho (NVIDIA), Marco Pavone (NVIDIA), Song Han (NVIDIA), Hongxu Yin (NVIDIA)
arXiv

Revisiting Semi-Supervised Learning in the Era of Foundation Models

Authors: Ping Zhang (OSU), Zheda Mai(OSU), Quang-Huy Nguyen(OSU), Wei-Lun Chao(OSU)
arXiv