1.8k citations · 1.9k across the 18 of their papers we have counts for
19 papers
EigDiff: Eigensample Diffusion - or how to significantly reduce training dataset size with marginal performance variation?
Andrei Jarca, Vlad Hondru, Florinel-Alin Croitoru +1
Diffusion models are notoriously expensive to train, requiring large-scale datasets to learn how to gradually reverse the noise from images over multiple small denoising steps. Cur…
DF-MoE: Generalizable Deepfake Detection via Multimodal Sparse Mixture-of-Experts
Vlad Hondru, Florinel Alin Croitoru, Iuliana Georgescu +2
Audio-visual deepfake detection is an actively studied topic, where one of the main challenges is to develop detectors able to generalize across deepfake generation methods. We con…
Few-Medoids: An Embarrassingly Simple Coreset Selection Method for Few-Shot Knowledge Distillation
Cemil-Andrei Dilmac, Florinel-Alin Croitoru, Radu Tudor Ionescu
Coreset selection aims to identify a small and highly representative subset of a massive dataset for efficient model training. The problem remains challenging even in the few-shot…
MTL-MAD: Multi-Task Learners are Effective Medical Anomaly Detectors
Bogdan Alexandru Bercean, Florinel Alin Croitoru, Vlad Hondru +3
Anomaly detection in medical images is a challenging task, since anomalies are not typically available during training. Recent methods leverage a single pretext task coupled with a…
Curriculum-DPO++: Direct Preference Optimization via Data and Model Curricula for Text-to-Image Generation
Florinel-Alin Croitoru, Vlad Hondru, Radu Tudor Ionescu +2
Direct Preference Optimization (DPO) has been proposed as an effective and efficient alternative to reinforcement learning from human feedback (RLHF). However, neither RLHF nor DPO…
PRNU-Bench: A Novel Benchmark and Model for PRNU-Based Camera Identification
Florinel Alin Croitoru, Vlad Hondru, Radu Tudor Ionescu
We propose a novel benchmark for camera identification via Photo Response Non-Uniformity (PRNU) estimation. The benchmark comprises 13K photos taken with 120+ cameras, where the tr…
MAVOS-DD: Multilingual Audio-Video Open-Set Deepfake Detection Benchmark
Florinel-Alin Croitoru, Vlad Hondru, Marius Popescu +3
We present the first large-scale open-set benchmark for multilingual audio-video deepfake detection. Our dataset comprises nearly 350 hours of real and fake videos across eight lan…
Task-Informed Anti-Curriculum by Masking Improves Downstream Performance on Text
Andrei Jarca, Florinel Alin Croitoru, Radu Tudor Ionescu
Masked language modeling has become a widely adopted unsupervised technique to pre-train large language models (LLMs). However, the process of selecting tokens for masking is rando…
DLCR: A Generative Data Expansion Framework via Diffusion for Clothes-Changing Person Re-ID
Nyle Siddiqui, Florinel Alin Croitoru, Gaurav Kumar Nayak +2
With the recent exhibited strength of generative diffusion models, an open research question is if images generated by these models can be used to learn better visual representatio…
Deepfake Media Generation and Detection in the Generative AI Era: A Survey and Outlook
Florinel-Alin Croitoru, Andrei-Iulian Hiji, Vlad Hondru +7
We survey deepfake generation and detection techniques, covering all deepfake media types: image, video, audio and multimodal content. We identify various kinds of deepfakes and co…
Masked Image Modeling: A Survey
Vlad Hondru, Florinel Alin Croitoru, Shervin Minaee +2
In this work, we survey recent studies on masked image modeling (MIM), an approach that emerged as a powerful self-supervised learning technique in computer vision. The MIM task in…
CBM: Curriculum by Masking
Andrei Jarca, Florinel-Alin Croitoru, Radu Tudor Ionescu
We propose Curriculum by Masking (CBM), a novel state-of-the-art curriculum learning strategy that effectively creates an easy-to-hard training schedule via patch (token) masking,…
Curriculum Direct Preference Optimization for Diffusion and Consistency Models
Florinel-Alin Croitoru, Vlad Hondru, Radu Tudor Ionescu +2
Direct Preference Optimization (DPO) has been proposed as an effective and efficient alternative to reinforcement learning from human feedback (RLHF). In this paper, we propose a n…
Reverse Stable Diffusion: What prompt was used to generate this image?
Florinel-Alin Croitoru, Vlad Hondru, Radu Tudor Ionescu +1
Text-to-image diffusion models have recently attracted the interest of many researchers, and inverting the diffusion process can play an important role in better understanding the…
Self-Distilled Masked Auto-Encoders are Efficient Video Anomaly Detectors
Nicolae-Catalin Ristea, Florinel-Alin Croitoru, Radu Tudor Ionescu +3
We propose an efficient abnormal event detection model based on a lightweight masked auto-encoder (AE) applied at the video frame level. The novelty of the proposed model is threef…
Lightning Fast Video Anomaly Detection via Adversarial Knowledge Distillation
Florinel-Alin Croitoru, Nicolae-Catalin Ristea, Dana Dascalescu +3
We propose a very fast frame-level model for anomaly detection in video, which learns to detect anomalies by distilling knowledge from multiple highly accurate object-level teacher…
Diffusion Models in Vision: A Survey
Florinel-Alin Croitoru, Vlad Hondru, Radu Tudor Ionescu +1
Denoising diffusion models represent a recent emerging topic in computer vision, demonstrating remarkable results in the area of generative modeling. A diffusion model is a deep ge…
Learning Rate Curriculum
Florinel-Alin Croitoru, Nicolae-Catalin Ristea, Radu Tudor Ionescu +1
Most curriculum learning methods require an approach to sort the data samples by difficulty, which is often cumbersome to perform. In this work, we propose a novel curriculum learn…
Discriminability-enforcing loss to improve representation learning
Florinel-Alin Croitoru, Diana-Nicoleta Grigore, Radu Tudor Ionescu
During the training process, deep neural networks implicitly learn to represent the input data samples through a hierarchy of features, where the size of the hierarchy is determine…