Ioannis Patras

Simone Barattin, Christos Tzelepis, Ioannis Patras et al.

This work addresses the problem of anonymizing the identity of faces in a dataset of images, such that the privacy of those depicted is not violated, while at the same time the dataset is useful for downstream task such as for training machine learning models. To the best of our knowledge, we are the first to explicitly address this issue and deal with two major drawbacks of the existing state-of-the-art approaches, namely that they (i) require the costly training of additional, purpose-trained neural networks, and/or (ii) fail to retain the facial attributes of the original images in the anonymized counterparts, the preservation of which is of paramount importance for their use in downstream tasks. We accordingly present a task-agnostic anonymization procedure that directly optimizes the images' latent representation in the latent space of a pre-trained GAN. By optimizing the latent codes directly, we ensure both that the identity is of a desired distance away from the original (with an identity obfuscation loss), whilst preserving the facial attributes (using a novel feature-matching loss in FaRL's deep feature space). We demonstrate through a series of both qualitative and quantitative experiments that our method is capable of anonymizing the identity of the images whilst -- crucially -- better-preserving the facial attributes. We make the code and the pre-trained models publicly available at: https://github.com/chi0tzp/FALCO.

Stella Bounareli, Christos Tzelepis, Vasileios Argyriou et al.

In this paper, we present our method for neural face reenactment, called HyperReenact, that aims to generate realistic talking head images of a source identity, driven by a target facial pose. Existing state-of-the-art face reenactment methods train controllable generative models that learn to synthesize realistic facial images, yet producing reenacted faces that are prone to significant visual artifacts, especially under the challenging condition of extreme head pose changes, or requiring expensive few-shot fine-tuning to better preserve the source identity characteristics. We propose to address these limitations by leveraging the photorealistic generation ability and the disentangled properties of a pretrained StyleGAN2 generator, by first inverting the real images into its latent space and then using a hypernetwork to perform: (i) refinement of the source identity characteristics and (ii) facial pose re-targeting, eliminating this way the dependence on external editing methods that typically produce artifacts. Our method operates under the one-shot setting (i.e., using a single source frame) and allows for cross-subject reenactment, without requiring any subject-specific fine-tuning. We compare our method both quantitatively and qualitatively against several state-of-the-art techniques on the standard benchmarks of VoxCeleb1 and VoxCeleb2, demonstrating the superiority of our approach in producing artifact-free images, exhibiting remarkable robustness even under extreme head pose changes. We make the code and the pretrained models publicly available at: https://github.com/StelaBou/HyperReenact .

Zengqun Zhao, Ioannis Patras

This paper presents a novel visual-language model called DFER-CLIP, which is based on the CLIP model and designed for in-the-wild Dynamic Facial Expression Recognition (DFER). Specifically, the proposed DFER-CLIP consists of a visual part and a textual part. For the visual part, based on the CLIP image encoder, a temporal model consisting of several Transformer encoders is introduced for extracting temporal facial expression features, and the final feature embedding is obtained as a learnable "class" token. For the textual part, we use as inputs textual descriptions of the facial behaviour that is related to the classes (facial expressions) that we are interested in recognising -- those descriptions are generated using large language models, like ChatGPT. This, in contrast to works that use only the class names and more accurately captures the relationship between them. Alongside the textual description, we introduce a learnable token which helps the model learn relevant context information for each expression during training. Extensive experiments demonstrate the effectiveness of the proposed method and show that our DFER-CLIP also achieves state-of-the-art results compared with the current supervised DFER methods on the DFEW, FERV39k, and MAFW benchmarks. Code is publicly available at https://github.com/zengqunzhao/DFER-CLIP.

James Oldfield, Christos Tzelepis, Yannis Panagakis et al.

Recent advances in the understanding of Generative Adversarial Networks (GANs) have led to remarkable progress in visual editing and synthesis tasks, capitalizing on the rich semantics that are embedded in the latent spaces of pre-trained GANs. However, existing methods are often tailored to specific GAN architectures and are limited to either discovering global semantic directions that do not facilitate localized control, or require some form of supervision through manually provided regions or segmentation masks. In this light, we present an architecture-agnostic approach that jointly discovers factors representing spatial parts and their appearances in an entirely unsupervised fashion. These factors are obtained by applying a semi-nonnegative tensor factorization on the feature maps, which in turn enables context-aware local image editing with pixel-level control. In addition, we show that the discovered appearance factors correspond to saliency maps that localize concepts of interest, without using any labels. Experiments on a wide range of GAN architectures and datasets show that, in comparison to the state of the art, our method is far more efficient in terms of training time and, most importantly, provides much more accurate localized control. Our code is available at: https://github.com/james-oldfield/PandA.

Chen Feng, Ioannis Patras

Self-supervised learning has recently achieved great success in representation learning without human annotations. The dominant method -- that is contrastive learning, is generally based on instance discrimination tasks, i.e., individual samples are treated as independent categories. However, presuming all the samples are different contradicts the natural grouping of similar samples in common visual datasets, e.g., multiple views of the same dog. To bridge the gap, this paper proposes an adaptive method that introduces soft inter-sample relations, namely Adaptive Soft Contrastive Learning (ASCL). More specifically, ASCL transforms the original instance discrimination task into a multi-instance soft discrimination task, and adaptively introduces inter-sample relations. As an effective and concise plug-in module for existing self-supervised learning frameworks, ASCL achieves the best performance on several benchmarks in terms of both performance and efficiency. Code is available at https://github.com/MrChenFeng/ASCL_ICPR2022.

Stella Bounareli, Christos Tzelepis, Vasileios Argyriou et al.

In this paper we address the problem of neural face reenactment, where, given a pair of a source and a target facial image, we need to transfer the target's pose (defined as the head pose and its facial expressions) to the source image, by preserving at the same time the source's identity characteristics (e.g., facial shape, hair style, etc), even in the challenging case where the source and the target faces belong to different identities. In doing so, we address some of the limitations of the state-of-the-art works, namely, a) that they depend on paired training data (i.e., source and target faces have the same identity), b) that they rely on labeled data during inference, and c) that they do not preserve identity in large head pose changes. More specifically, we propose a framework that, using unpaired randomly generated facial images, learns to disentangle the identity characteristics of the face from its pose by incorporating the recently introduced style space $\mathcal{S}$ of StyleGAN2, a latent representation space that exhibits remarkable disentanglement properties. By capitalizing on this, we learn to successfully mix a pair of source and target style codes using supervision from a 3D model. The resulting latent code, that is subsequently used for reenactment, consists of latent units corresponding to the facial pose of the target only and of units corresponding to the identity of the source only, leading to notable improvement in the reenactment performance compared to recent state-of-the-art methods. In comparison to state of the art, we quantitatively and qualitatively show that the proposed method produces higher quality results even on extreme pose variations. Finally, we report results on real images by first embedding them on the latent space of the pretrained generator. We make the code and pretrained models publicly available at: https://github.com/StelaBou/StyleMask

Chen Feng, Ioannis Patras

Deep learning has achieved great success in recent years with the aid of advanced neural network structures and large-scale human-annotated datasets. However, it is often costly and difficult to accurately and efficiently annotate large-scale datasets, especially for some specialized domains where fine-grained labels are required. In this setting, coarse labels are much easier to acquire as they do not require expert knowledge. In this work, we propose a contrastive learning method, called $\textbf{Mask}$ed $\textbf{Con}$trastive learning~($\textbf{MaskCon}$) to address the under-explored problem setting, where we learn with a coarse-labelled dataset in order to address a finer labelling problem. More specifically, within the contrastive learning framework, for each sample our method generates soft-labels with the aid of coarse labels against other samples and another augmented view of the sample in question. By contrast to self-supervised contrastive learning where only the sample's augmentations are considered hard positives, and in supervised contrastive learning where only samples with the same coarse labels are considered hard positives, we propose soft labels based on sample distances, that are masked by the coarse labels. This allows us to utilize both inter-sample relations and coarse labels. We demonstrate that our method can obtain as special cases many existing state-of-the-art works and that it provides tighter bounds on the generalization error. Experimentally, our method achieves significant improvement over the current state-of-the-art in various datasets, including CIFAR10, CIFAR100, ImageNet-1K, Standford Online Products and Stanford Cars196 datasets. Code and annotations are available at https://github.com/MrChenFeng/MaskCon_CVPR2023.

Christos Tzelepis, James Oldfield, Georgios Tzimiropoulos et al.

This work addresses the problem of discovering non-linear interpretable paths in the latent space of pre-trained GANs in a model-agnostic manner. In the proposed method, the discovery is driven by a set of pairs of natural language sentences with contrasting semantics, named semantic dipoles, that serve as the limits of the interpretation that we require by the trainable latent paths to encode. By using the pre-trained CLIP encoder, the sentences are projected into the vision-language space, where they serve as dipoles, and where RBF-based warping functions define a set of non-linear directional paths, one for each semantic dipole, allowing in this way traversals from one semantic pole to the other. By defining an objective that discovers paths in the latent space of GANs that generate changes along the desired paths in the vision-language embedding space, we provide an intuitive way of controlling the underlying generative factors and address some of the limitations of the state-of-the-art works, namely, that a) they are typically tailored to specific GAN architectures (i.e., StyleGAN), b) they disregard the relative position of the manipulated and the original image in the image embedding and the relative position of the image and the text embeddings, and c) they lead to abrupt image manipulations and quickly arrive at regions of low density and, thus, low image quality, providing limited control of the generative factors. We provide extensive qualitative and quantitative results that demonstrate our claims with two pre-trained GANs, and make the code and the pre-trained models publicly available at: https://github.com/chi0tzp/ContraCLIP

Giorgos Kordopatis-Zilos, Giorgos Tolias, Christos Tzelepis et al.

We introduce S$^2$VS, a video similarity learning approach with self-supervision. Self-Supervised Learning (SSL) is typically used to train deep models on a proxy task so as to have strong transferability on target tasks after fine-tuning. Here, in contrast to prior work, SSL is used to perform video similarity learning and address multiple retrieval and detection tasks at once with no use of labeled data. This is achieved by learning via instance-discrimination with task-tailored augmentations and the widely used InfoNCE loss together with an additional loss operating jointly on self-similarity and hard-negative similarity. We benchmark our method on tasks where video relevance is defined with varying granularity, ranging from video copies to videos depicting the same incident or event. We learn a single universal model that achieves state-of-the-art performance on all tasks, surpassing previously proposed methods that use labeled data. The code and pretrained models are publicly available at: https://github.com/gkordo/s2vs

Georgios Zoumpourlis, Ioannis Patras

In this work, we study the problem of cross-subject motor imagery (MI) decoding from electroencephalography (EEG) data. Multi-subject EEG datasets present several kinds of domain shifts due to various inter-individual differences (e.g. brain anatomy, personality and cognitive profile). These domain shifts render multi-subject training a challenging task and also impede robust cross-subject generalization. Inspired by the importance of domain generalization techniques for tackling such issues, we propose a two-stage model ensemble architecture built with multiple feature extractors (first stage) and a shared classifier (second stage), which we train end-to-end with two novel loss terms. The first loss applies curriculum learning, forcing each feature extractor to specialize to a subset of the training subjects and promoting feature diversity. The second loss is an intra-ensemble distillation objective that allows collaborative exchange of knowledge between the models of the ensemble. We compare our method against several state-of-the-art techniques, conducting subject-independent experiments on two large MI datasets, namely PhysioNet and OpenBMI. Our algorithm outperforms all of the methods in both 5-fold cross-validation and leave-one-subject-out evaluation settings, using a substantially lower number of trainable parameters. We demonstrate that our model ensembling approach combining the powers of curriculum learning and collaborative training, leads to high learning capacity and robust performance. Our work addresses the issue of domain shifts in multi-subject EEG datasets, paving the way for calibration-free brain-computer interfaces. We make our code publicly available at: https://github.com/gzoumpourlis/Ensemble-MI

Alexandros Xenos, Themos Stafylakis, Ioannis Patras et al.

This paper is on the problem of Knowledge-Based Visual Question Answering (KB-VQA). Recent works have emphasized the significance of incorporating both explicit (through external databases) and implicit (through LLMs) knowledge to answer questions requiring external knowledge effectively. A common limitation of such approaches is that they consist of relatively complicated pipelines and often heavily rely on accessing GPT-3 API. Our main contribution in this paper is to propose a much simpler and readily reproducible pipeline which, in a nutshell, is based on efficient in-context learning by prompting LLaMA (1 and 2) using question-informative captions as contextual information. Contrary to recent approaches, our method is training-free, does not require access to external databases or APIs, and yet achieves state-of-the-art accuracy on the OK-VQA and A-OK-VQA datasets. Finally, we perform several ablation studies to understand important aspects of our method. Our code is publicly available at https://github.com/alexandrosXe/ASimple-Baseline-For-Knowledge-Based-VQA

Moreno D'Incà, Christos Tzelepis, Ioannis Patras et al.

Fairness is crucial when training a deep-learning discriminative model, especially in the facial domain. Models tend to correlate specific characteristics (such as age and skin color) with unrelated attributes (downstream tasks), resulting in biases which do not correspond to reality. It is common knowledge that these correlations are present in the data and are then transferred to the models during training. This paper proposes a method to mitigate these correlations to improve fairness. To do so, we learn interpretable and meaningful paths lying in the semantic space of a pre-trained diffusion model (DiffAE) -- such paths being supervised by contrastive text dipoles. That is, we learn to edit protected characteristics (age and skin color). These paths are then applied to augment images to improve the fairness of a given dataset. We test the proposed method on CelebA-HQ and UTKFace on several downstream tasks with age and skin color as protected characteristics. As a proxy for fairness, we compute the difference in accuracy with respect to the protected characteristics. Quantitative results show how the augmented images help the model improve the overall accuracy, the aforementioned metric, and the disparity of equal opportunity. Code is available at: https://github.com/Moreno98/Vision-Language-Bias-Control.

Niki Maria Foteinopoulou, Ioannis Patras

Facial Expression Recognition (FER) is a crucial task in affective computing, but its conventional focus on the seven basic emotions limits its applicability to the complex and expanding emotional spectrum. To address the issue of new and unseen emotions present in dynamic in-the-wild FER, we propose a novel vision-language model that utilises sample-level text descriptions (i.e. captions of the context, expressions or emotional cues) as natural language supervision, aiming to enhance the learning of rich latent representations, for zero-shot classification. To test this, we evaluate using zero-shot classification of the model trained on sample-level descriptions on four popular dynamic FER datasets. Our findings show that this approach yields significant improvements when compared to baseline methods. Specifically, for zero-shot video FER, we outperform CLIP by over 10\% in terms of Weighted Average Recall and 5\% in terms of Unweighted Average Recall on several datasets. Furthermore, we evaluate the representations obtained from the network trained using sample-level descriptions on the downstream task of mental health symptom estimation, achieving performance comparable or superior to state-of-the-art methods and strong agreement with human experts. Namely, we achieve a Pearson's Correlation Coefficient of up to 0.85 on schizophrenia symptom severity estimation, which is comparable to human experts' agreement. The code is publicly available at: https://github.com/NickyFot/EmoCLIP.

Debin Meng, Christos Tzelepis, Ioannis Patras et al.

Generating human portraits is a hot topic in the image generation area, e.g. mask-to-face generation and text-to-face generation. However, these unimodal generation methods lack controllability in image generation. Controllability can be enhanced by exploring the advantages and complementarities of various modalities. For instance, we can utilize the advantages of text in controlling diverse attributes and masks in controlling spatial locations. Current state-of-the-art methods in multimodal generation face limitations due to their reliance on extensive hyperparameters, manual operations during the inference stage, substantial computational demands during training and inference, or inability to edit real images. In this paper, we propose a practical framework - MM2Latent - for multimodal image generation and editing. We use StyleGAN2 as our image generator, FaRL for text encoding, and train an autoencoders for spatial modalities like mask, sketch and 3DMM. We propose a strategy that involves training a mapping network to map the multimodal input into the w latent space of StyleGAN. The proposed framework 1) eliminates hyperparameters and manual operations in the inference stage, 2) ensures fast inference speeds, and 3) enables the editing of real images. Extensive experiments demonstrate that our method exhibits superior performance in multimodal image generation, surpassing recent GAN- and diffusion-based methods. Also, it proves effective in multimodal image editing and is faster than GAN- and diffusion-based methods. We make the code publicly available at: https://github.com/Open-Debin/MM2Latent

Dario Cioni, Christos Tzelepis, Lorenzo Seidenari et al.

The steady improvement of Diffusion Models for visual synthesis has given rise to many new and interesting use cases of synthetic images but also has raised concerns about their potential abuse, which poses significant societal threats. To address this, fake images need to be detected and attributed to their source model, and given the frequent release of new generators, realistic applications need to consider an Open-Set scenario where some models are unseen at training time. Existing forensic techniques are either limited to Closed-Set settings or to GAN-generated images, relying on fragile frequency-based "fingerprint" features. By contrast, we propose a simple yet effective framework that incorporates features from large pre-trained foundation models to perform Open-Set origin attribution of synthetic images produced by various generative models, including Diffusion Models. We show that our method leads to remarkable attribution performance, even in the low-data regime, exceeding the performance of existing methods and generalizes better on images obtained from a diverse set of architectures. We make the code publicly available at: https://github.com/ciodar/UniversalAttribution.

Ioannis Maniadis Metaxas, Georgios Tzimiropoulos, Ioannis Patras

Clustering has been a major research topic in the field of machine learning, one to which Deep Learning has recently been applied with significant success. However, an aspect of clustering that is not addressed by existing deep clustering methods, is that of efficiently producing multiple, diverse partitionings for a given dataset. This is particularly important, as a diverse set of base clusterings are necessary for consensus clustering, which has been found to produce better and more robust results than relying on a single clustering. To address this gap, we propose DivClust, a diversity controlling loss that can be incorporated into existing deep clustering frameworks to produce multiple clusterings with the desired degree of diversity. We conduct experiments with multiple datasets and deep clustering frameworks and show that: a) our method effectively controls diversity across frameworks and datasets with very small additional computational cost, b) the sets of clusterings learned by DivClust include solutions that significantly outperform single-clustering baselines, and c) using an off-the-shelf consensus clustering algorithm, DivClust produces consensus clustering solutions that consistently outperform single-clustering baselines, effectively improving the performance of the base deep clustering framework.

Niki Maria Foteinopoulou, Ioannis Patras

Human affect and mental state estimation in an automated manner, face a number of difficulties, including learning from labels with poor or no temporal resolution, learning from few datasets with little data (often due to confidentiality constraints) and, (very) long, in-the-wild videos. For these reasons, deep learning methodologies tend to overfit, that is, arrive at latent representations with poor generalisation performance on the final regression task. To overcome this, in this work, we introduce two complementary contributions. First, we introduce a novel relational loss for multilabel regression and ordinal problems that regularises learning and leads to better generalisation. The proposed loss uses label vector inter-relational information to learn better latent representations by aligning batch label distances to the distances in the latent feature space. Second, we utilise a two-stage attention architecture that estimates a target for each clip by using features from the neighbouring clips as temporal context. We evaluate the proposed methodology on both continuous affect and schizophrenia severity estimation problems, as there are methodological and contextual parallels between the two. Experimental results demonstrate that the proposed methodology outperforms all baselines. In the domain of schizophrenia, the proposed methodology outperforms previous state-of-the-art by a large margin, achieving a PCC of up to 78% performance close to that of human experts (85%) and much higher than previous works (uplift of up to 40%). In the case of affect recognition, we outperform previous vision-based methods in terms of CCC on both the OMG and the AMIGOS datasets. Specifically for AMIGOS, we outperform previous SoTA CCC for both arousal and valence by 9% and 13% respectively, and in the OMG dataset we outperform previous vision works by up to 5% for both arousal and valence.

Zengqun Zhao, Ziquan Liu, Yu Cao et al.

The recent success of inference-time scaling in large language models has inspired similar explorations in video diffusion. In particular, motivated by the existence of "golden noise" that enhances video quality, prior work has attempted to improve inference by optimising or searching for better initial noise. However, these approaches have notable limitations: they either rely on priors imposed at the beginning of noise sampling or on rewards evaluated only on the denoised and decoded videos. This leads to error accumulation, delayed and sparse reward signals, and prohibitive computational cost, which prevents the use of stronger search algorithms. Crucially, stronger search algorithms are precisely what could unlock substantial gains in controllability, sample efficiency and generation quality for video diffusion, provided their computational cost can be reduced. To fill in this gap, we enable efficient inference-time scaling for video diffusion through latent reward guidance, which provides intermediate, informative and efficient feedback along the denoising trajectory. We introduce a latent reward model that scores partially denoised latents at arbitrary timesteps with respect to visual quality, motion quality, and text alignment. Building on this model, we propose LatSearch, a novel inference-time search mechanism that performs Reward-Guided Resampling and Pruning (RGRP). In the resampling stage, candidates are sampled according to reward-normalised probabilities to reduce over-reliance on the reward model. In the pruning stage, applied at the final scheduled step, only the candidate with the highest cumulative reward is retained, improving both quality and efficiency. We evaluate LatSearch on the VBench-2.0 benchmark and demonstrate that it consistently improves video generation across multiple evaluation dimensions compared to the baseline Wan2.1 model.

Ioannis Maniadis Metaxas, Adrian Bulat, Ioannis Patras et al.

The unsupervised pretraining of object detectors has recently become a key component of object detector training, as it leads to improved performance and faster convergence during the supervised fine-tuning stage. Existing unsupervised pretraining methods, however, typically rely on low-level information to define proposals that are used to train the detector. Furthermore, in the absence of class labels for these proposals, an auxiliary loss is used to add high-level semantics. This results in complex pipelines and a task gap between the pretraining and the downstream task. We propose a framework that mitigates this issue and consists of three simple yet key ingredients: (i) richer initial proposals that do encode high-level semantics, (ii) class pseudo-labeling through clustering, that enables pretraining using a standard object detection training pipeline, (iii) self-training to iteratively improve and enrich the object proposals. Once the pretraining and downstream tasks are aligned, a simple detection pipeline without further bells and whistles can be directly used for pretraining and, in fact, results in state-of-the-art performance on both the full and low data regimes, across detector architectures and datasets, by significant margins. We further show that our pretraining strategy is also capable of pretraining from scratch (including the backbone) and works on complex images like COCO, paving the path for unsupervised representation learning using object detection directly as a pretext task.

Chen Feng, Georgios Tzimiropoulos, Ioannis Patras

Learning with Noisy labels (LNL) poses a significant challenge for the Machine Learning community. Some of the most widely used approaches that select as clean samples for which the model itself (the in-training model) has high confidence, e.g., `small loss', can suffer from the so called `self-confirmation' bias. This bias arises because the in-training model, is at least partially trained on the noisy labels. Furthermore, in the classification case, an additional challenge arises because some of the label noise is between classes that are visually very similar (`hard noise'). This paper addresses these challenges by proposing a method (\textit{CLIPCleaner}) that leverages CLIP, a powerful Vision-Language (VL) model for constructing a zero-shot classifier for efficient, offline, clean sample selection. This has the advantage that the sample selection is decoupled from the in-training model and that the sample selection is aware of the semantic and visual similarities between the classes due to the way that CLIP is trained. We provide theoretical justifications and empirical evidence to demonstrate the advantages of CLIP for LNL compared to conventional pre-trained models. Compared to current methods that combine iterative sample selection with various techniques, \textit{CLIPCleaner} offers a simple, single-step approach that achieves competitive or superior performance on benchmark datasets. To the best of our knowledge, this is the first time a VL model has been used for sample selection to address the problem of Learning with Noisy Labels (LNL), highlighting their potential in the domain.

Zhonglin Sun, Siyang Song, Ioannis Patras et al.

Privacy issue is a main concern in developing face recognition techniques. Although synthetic face images can partially mitigate potential legal risks while maintaining effective face recognition (FR) performance, FR models trained by face images synthesized by existing generative approaches frequently suffer from performance degradation problems due to the insufficient discriminative quality of these synthesized samples. In this paper, we systematically investigate what contributes to solid face recognition model training, and reveal that face images with certain degree of similarities to their identity centers show great effectiveness in the performance of trained FR models. Inspired by this, we propose a novel diffusion-based approach (namely Center-based Semi-hard Synthetic Face Generation (CemiFace)) which produces facial samples with various levels of similarity to the subject center, thus allowing to generate face datasets containing effective discriminative samples for training face recognition. Experimental results show that with a modest degree of similarity, training on the generated dataset can produce competitive performance compared to previous generation methods.

Yuangong Chen, Wai Keung Wong, Jiaxing Li et al.

Multimodal Large Language Models (MLLMs) show strong visual perception, yet remain limited in reasoning about space under changing viewpoints. We study this challenge as Perspective-Conditioned Spatial Reasoning (PCSR) in 360-degree omnidirectional images, where broad scene coverage reduces ambiguity from partial observations without eliminating the need for viewpoint-dependent inference. To assess this capability, we introduce PCSR-Bench, a diagnostic benchmark of 84,373 question-answer pairs from 2,600 omnidirectional images across 26 indoor environments. PCSR-Bench contains eight tasks spanning foundational perception (e.g., object counting, relative distance, and relative direction) and advanced PCSR, including compositional chains, egocentric rotation, perspective re-anchoring, ego-distortion, and limited-FOV visibility. We evaluate 14 representative MLLMs and observe a substantial perception-reasoning gap: accuracy reaches 57.59% on foundational relative direction, but drops to 13.49% on egocentric rotation, 7.13% on egocentric distortion, and 0.64% on open-ended compositional reasoning. To probe the plasticity of this gap, we conduct an RL-based diagnostic study on a 7B-scale model. Reward shaping improves a matched 7B baseline from 31.10% to 60.06% under a controlled setting, suggesting that PCSR is partial plasticity rather than being fully immutable. Still, the gains are task-selective, sensitive to reward design including both weight allocation and reward formulation, and partially dependent on the evaluation protocol. These results position PCSR as a key bottleneck in current MLLMs and highlight limited but meaningful room for recovery under targeted optimization.

Harsh Panwar, Ioannis Patras

Capsule Networks have shown tremendous advancement in the past decade, outperforming the traditional CNNs in various task due to it's equivariant properties. With the use of vector I/O which provides information of both magnitude and direction of an object or it's part, there lies an enormous possibility of using Capsule Networks in unsupervised learning environment for visual representation tasks such as multi class image classification. In this paper, we propose Contrastive Capsule (CoCa) Model which is a Siamese style Capsule Network using Contrastive loss with our novel architecture, training and testing algorithm. We evaluate the model on unsupervised image classification CIFAR-10 dataset and achieve a top-1 test accuracy of 70.50% and top-5 test accuracy of 98.10%. Due to our efficient architecture our model has 31 times less parameters and 71 times less FLOPs than the current SOTA in both supervised and unsupervised learning.

Jiawei Ge, Xintian Zhang, Jiuxin Cao et al.

Cross-view Referring Multi-Object Tracking (CRMOT) aims to track multiple objects specified by natural language across multiple camera views, with globally consistent identities. Despite recent progress, existing methods rely heavily on costly frame-level spatial annotations and cross-view identity supervision. To reduce such reliance, we explore CRMOT under weak supervision by leveraging the capabilities of foundation models. However, our empirical study shows that directly applying foundation models such as SAM2 and SAM3, even with task-specific modifications, fails to accurately understand referring expressions and maintain consistent identities across views. Yet, they remain effective at producing reliable object tracklets that can serve as pseudo supervision. We therefore repurpose foundation models as pseudo-label generators and propose a two-stage framework for weakly supervised CRMOT, using only object category labels as coarse-grained supervision. In the first stage, we design an Affinity-guided Cross-view Re-prompting strategy to refine and associate SAM3-generated tracklets across cameras, producing reliable cross-view pseudo labels for subsequent training. In the second stage, we introduce ViewSAM, a CRMOT model built upon SAM2 that explicitly models view-aware cross-modal semantics. By formulating view-induced variations as learnable conditions, ViewSAM bridges the gap between view-variant visual observations and view-invariant textual expressions, enabling robust cross-view referring tracking with only approximately 10% additional parameters. Extensive experiments demonstrate that ViewSAM achieves SOTA performance under weak supervision and remains competitive with fully supervised methods.

Zhaohan Zhang, Chengzhengxu Li, Xiaoming Liu et al.

Large Language Models (LLMs) are increasingly applied in high-stakes domains such as finance, healthcare, and education, where reliable multi-turn interactions with users are essential. However, existing work on confidence estimation and calibration, a major approach to building trustworthy LLM systems, largely focuses on single-turn settings and overlooks the risks and potential of multi-turn conversations. In this work, we introduce the task of multi-turn calibration to reframe calibration from a static property into a dynamic challenge central to reliable multi-turn conversation, where calibrating model confidence at each turn conditioned on the conversation history is required. We first reveal the risks of this setting: using Expected Calibration Error at turn T (ECE@T), a new metric that tracks calibration dynamics over turns, we show that user feedback (e.g., persuasion) can degrade multi-turn calibration. To address this, we propose MTCal, which minimises ECE@T via a surrogate calibration target, and further leverage calibrated confidence in ConfChat, a decoding strategy that improves both factuality and consistency of the model response in multi-turn interactions. Extensive experiments demonstrate that MT-Cal achieves outstanding and consistent performance in multi-turn calibration, and ConfChat preserves and even enhances model performance in multi-turn interactions. Our results mark multi-turn calibration as one missing link for scaling LLM calibration toward safe, reliable, and real-world use.

Marios Krestenitis, Christos Tzelepis, Konstantinos Ioannidis et al.

Visual-Language Models (VLMs) have achieved remarkable progress in image captioning, visual question answering, and visual reasoning. Yet they remain prone to vision-language misalignment, often producing overly generic or hallucinated descriptions. Existing approaches address this via instruction tuning-requiring costly, large-scale annotated datasets or via complex test-time frameworks for caption refinement. In this work, we revisit image-text alignment through the lens of cycle consistency: given an image and a caption generated by an image-to-text model, the backward mapping through a text-to-image model should reconstruct an image that closely matches the original. In our setup, a VLM serves as the image-to-text component, while a pre-trained text-to-image model closes the loop by reconstructing the image from the generated caption. Building on this, we introduce CycleCap, a fine-tuning scheme to improve image captioning using Group Relative Policy Optimization (GRPO) with a reward based on the similarity between the original and reconstructed images, computed on-the-fly. Unlike previous work that uses cycle consistency loss for preference dataset construction, our method leverages cycle consistency directly as a self-supervised training signal. This enables the use of raw images alone, eliminating the need for curated image-text datasets, while steering the VLM to produce more accurate and grounded text descriptions. Applied to four VLMs ranging from 1B to 7B parameters, CycleCap yields consistent improvements across captioning and hallucination benchmarks, surpassing state-of-the-art methods that rely on supervised cycle consistency training.

Ioannis Maniadis Metaxas, Georgios Tzimiropoulos, Ioannis Patras

Self-supervised learning has recently emerged as the preeminent pretraining paradigm across and between modalities, with remarkable results. In the image domain specifically, group (or cluster) discrimination has been one of the most successful methods. However, such frameworks need to guard against heavily imbalanced cluster assignments to prevent collapse to trivial solutions. Existing works typically solve this by reweighing cluster assignments to promote balance, or with offline operations (e.g. regular re-clustering) that prevent collapse. However, the former typically requires large batch sizes, which leads to increased resource requirements, and the latter introduces scalability issues with regard to large datasets. In this work, we propose ExCB, a framework that tackles this problem with a novel cluster balancing method. ExCB estimates the relative size of the clusters across batches and balances them by adjusting cluster assignments, proportionately to their relative size and in an online manner. Thereby, it overcomes previous methods' dependence on large batch sizes and is fully online, and therefore scalable to any dataset. We conduct extensive experiments to evaluate our approach and demonstrate that ExCB: a) achieves state-of-the-art results with significantly reduced resource requirements compared to previous works, b) is fully online, and therefore scalable to large datasets, and c) is stable and effective even with very small batch sizes.

Zheng Gao, Chen Feng, Ioannis Patras

Whilst contrastive learning yields powerful representations by matching different augmented views of the same instance, it lacks the ability to capture the similarities between different instances. One popular way to address this limitation is by learning global features (after the global pooling) to capture inter-instance relationships based on knowledge distillation, where the global features of the teacher are used to guide the learning of the global features of the student. Inspired by cross-modality learning, we extend this existing framework that only learns from global features by encouraging the global features and intermediate layer features to learn from each other. This leads to our novel self-supervised framework: cross-context learning between global and hypercolumn features (CGH), that enforces the consistency of instance relations between low- and high-level semantics. Specifically, we stack the intermediate feature maps to construct a hypercolumn representation so that we can measure instance relations using two contexts (hypercolumn and global feature) separately, and then use the relations of one context to guide the learning of the other. This cross-context learning allows the model to learn from the differences between the two contexts. The experimental results on linear classification and downstream tasks show that our method outperforms the state-of-the-art methods.

Alexandros Xenos, Niki Maria Foteinopoulou, Ioanna Ntinou et al.

Recognising emotions in context involves identifying an individual's apparent emotions while considering contextual cues from the surrounding scene. Previous approaches to this task have typically designed explicit scene-encoding architectures or incorporated external scene-related information, such as captions. However, these methods often utilise limited contextual information or rely on intricate training pipelines to decouple noise from relevant information. In this work, we leverage the capabilities of Vision-and-Large-Language Models (VLLMs) to enhance in-context emotion classification in a more straightforward manner. Our proposed method follows a simple yet effective two-stage approach. First, we prompt VLLMs to generate natural language descriptions of the subject's apparent emotion in relation to the visual context. Second, the descriptions, along with the visual input, are used to train a transformer-based architecture that fuses text and visual features before the final classification task. This method not only simplifies the training process but also significantly improves performance. Experimental results demonstrate that the textual descriptions effectively guide the model to constrain the noisy visual input, allowing our fused architecture to outperform individual modalities. Our approach achieves state-of-the-art performance across three datasets, BoLD, EMOTIC, and CAER-S, without bells and whistles. The code will be made publicly available on github: https://github.com/NickyFot/EmoCommonSense.git

James Oldfield, Markos Georgopoulos, Grigorios G. Chrysos et al.

The Mixture of Experts (MoE) paradigm provides a powerful way to decompose dense layers into smaller, modular computations often more amenable to human interpretation, debugging, and editability. However, a major challenge lies in the computational cost of scaling the number of experts high enough to achieve fine-grained specialization. In this paper, we propose the Multilinear Mixture of Experts ($μ$MoE) layer to address this, focusing on vision models. $μ$MoE layers enable scalable expert specialization by performing an implicit computation on prohibitively large weight tensors entirely in factorized form. Consequently, $μ$MoEs (1) avoid the restrictively high inference-time costs of dense MoEs, yet (2) do not inherit the training issues of the popular sparse MoEs' discrete (non-differentiable) expert routing. We present both qualitative and quantitative evidence that scaling $μ$MoE layers when fine-tuning foundation models for vision tasks leads to more specialized experts at the class-level, further enabling manual bias correction in CelebA attribute classification. Finally, we show qualitative results demonstrating the expert specialism achieved when pre-training large GPT2 and MLP-Mixer models with parameter-matched $μ$MoE blocks at every layer, maintaining comparable accuracy. Our code is available at: https://github.com/james-oldfield/muMoE.

Omnia Alwazzan, Ioannis Patras, Gregory Slabaugh

Fusion of multimodal healthcare data holds great promise to provide a holistic view of a patient's health, taking advantage of the complementarity of different modalities while leveraging their correlation. This paper proposes a simple and effective approach, inspired by attention, to fuse discriminative features from different modalities. We propose a novel attention mechanism, called Flattened Outer Arithmetic Attention (FOAA), which relies on outer arithmetic operators (addition, subtraction, product, and division) to compute attention scores from keys, queries and values derived from flattened embeddings of each modality. We demonstrate how FOAA can be implemented for self-attention and cross-attention, providing a reusable component in neural network architectures. We evaluate FOAA on two datasets for multimodal tumor classification and achieve state-of-the-art results, and we demonstrate that features enriched by FOAA are superior to those derived from other fusion approaches. The code is publicly available at \href{https://github.com/omniaalwazzan/FOAA}{https://github.com/omniaalwazzan/FOAA}

Simone Ricci, Niccolò Biondi, Federico Pernici et al.

Retrieval systems rely on representations learned by increasingly powerful models. However, due to the high training cost and inconsistencies in learned representations, there is significant interest in facilitating communication between representations and ensuring compatibility across independently trained neural networks. In the literature, two primary approaches are commonly used to adapt different learned representations: affine transformations, which adapt well to specific distributions but can significantly alter the original representation, and orthogonal transformations, which preserve the original structure with strict geometric constraints but limit adaptability. A key challenge is adapting the latent spaces of updated models to align with those of previous models on downstream distributions while preserving the newly learned representation spaces. In this paper, we impose a relaxed orthogonality constraint, namely $λ$-Orthogonality regularization, while learning an affine transformation, to obtain distribution-specific adaptation while retaining the original learned representations. Extensive experiments across various architectures and datasets validate our approach, demonstrating that it preserves the model's zero-shot performance and ensures compatibility across model updates. Code available at: \href{https://github.com/miccunifi/lambda_orthogonality.git}{https://github.com/miccunifi/lambda\_orthogonality}.

Andreas Goulas, Vasileios Mezaris, Ioannis Patras

To address computational and memory limitations of Large Multimodal Models in the Video Question-Answering task, several recent methods extract textual representations per frame (e.g., by captioning) and feed them to a Large Language Model (LLM) that processes them to produce the final response. However, in this way, the LLM does not have access to visual information and often has to process repetitive textual descriptions of nearby frames. To address those shortcomings, in this paper, we introduce VidCtx, a novel training-free VideoQA framework which integrates both modalities, i.e. both visual information from input frames and textual descriptions of others frames that give the appropriate context. More specifically, in the proposed framework a pre-trained Large Multimodal Model (LMM) is prompted to extract at regular intervals, question-aware textual descriptions (captions) of video frames. Those will be used as context when the same LMM will be prompted to answer the question at hand given as input a) a certain frame, b) the question and c) the context/caption of an appropriate frame. To avoid redundant information, we chose as context the descriptions of distant frames. Finally, a simple yet effective max pooling mechanism is used to aggregate the frame-level decisions. This methodology enables the model to focus on the relevant segments of the video and scale to a high number of frames. Experiments show that VidCtx achieves competitive performance among approaches that rely on open models on three public Video QA benchmarks, NExT-QA, IntentQA and STAR. Our code is available at https://github.com/IDT-ITI/VidCtx.

James Oldfield, Philip Torr, Ioannis Patras et al.

Monitoring large language models' (LLMs) activations is an effective way to detect harmful requests before they lead to unsafe outputs. However, traditional safety monitors often require the same amount of compute for every query. This creates a trade-off: expensive monitors waste resources on easy inputs, while cheap ones risk missing subtle cases. We argue that safety monitors should be flexible--costs should rise only when inputs are difficult to assess, or when more compute is available. To achieve this, we introduce Truncated Polynomial Classifiers (TPCs), a natural extension of linear probes for dynamic activation monitoring. Our key insight is that polynomials can be trained and evaluated progressively, term-by-term. At test-time, one can early-stop for lightweight monitoring, or use more terms for stronger guardrails when needed. TPCs provide two modes of use. First, as a safety dial: by evaluating more terms, developers and regulators can "buy" stronger guardrails from the same model. Second, as an adaptive cascade: clear cases exit early after low-order checks, and higher-order guardrails are evaluated only for ambiguous inputs, reducing overall monitoring costs. On two large-scale safety datasets (WildGuardMix and BeaverTails), for 4 models with up to 30B parameters, we show that TPCs compete with or outperform MLP-based probe baselines of the same size, all the while being more interpretable than their black-box counterparts. Our code is available at http://github.com/james-oldfield/tpc.

James Oldfield, Shawn Im, Sharon Li et al.

Multilayer perceptrons (MLPs) are an integral part of large language models, yet their dense representations render them difficult to understand, edit, and steer. Recent methods learn interpretable approximations via neuron-level sparsity, yet fail to faithfully reconstruct the original mapping--significantly increasing model's next-token cross-entropy loss. In this paper, we advocate for moving to layer-level sparsity to overcome the accuracy trade-off in sparse layer approximation. Under this paradigm, we introduce Mixture of Decoders (MxDs). MxDs generalize MLPs and Gated Linear Units, expanding pre-trained dense layers into tens of thousands of specialized sublayers. Through a flexible form of tensor factorization, each sparsely activating MxD sublayer implements a linear transformation with full-rank weights--preserving the original decoders' expressive capacity even under heavy sparsity. Experimentally, we show that MxDs significantly outperform state-of-the-art methods (e.g., Transcoders) on the sparsity-accuracy frontier in language models with up to 3B parameters. Further evaluations on sparse probing and feature steering demonstrate that MxDs learn similarly specialized features of natural language--opening up a promising new avenue for designing interpretable yet faithful decompositions. Our code is included at: https://github.com/james-oldfield/MxD/.

Chen Feng, Georgios Tzimiropoulos, Ioannis Patras

Despite the large progress in supervised learning with neural networks, there are significant challenges in obtaining high-quality, large-scale and accurately labelled datasets. In such a context, how to learn in the presence of noisy labels has received more and more attention. As a relatively complex problem, in order to achieve good results, current approaches often integrate components from several fields, such as supervised learning, semi-supervised learning, transfer learning and resulting in complicated methods. Furthermore, they often make multiple assumptions about the type of noise of the data. This affects the model robustness and limits its performance under different noise conditions. In this paper, we consider a novel problem setting, Learning with Unknown Label Noise}(LULN), that is, learning when both the degree and the type of noise are unknown. Under this setting, unlike previous methods that often introduce multiple assumptions and lead to complex solutions, we propose a simple, efficient and robust framework named Sample Selection and Relabelling(SSR), that with a minimal number of hyperparameters achieves SOTA results in various conditions. At the heart of our method is a sample selection and relabelling mechanism based on a non-parametric KNN classifier~(NPK) $g_q$ and a parametric model classifier~(PMC) $g_p$, respectively, to select the clean samples and gradually relabel the noisy samples. Without bells and whistles, such as model co-training, self-supervised pre-training and semi-supervised learning, and with robustness concerning the settings of its few hyper-parameters, our method significantly surpasses previous methods on both CIFAR10/CIFAR100 with synthetic noise and real-world noisy datasets such as WebVision, Clothing1M and ANIMAL-10N. Code is available at https://github.com/MrChenFeng/SSR_BMVC2022.

Christos Tzelepis, Georgios Tzimiropoulos, Ioannis Patras

This work addresses the problem of discovering, in an unsupervised manner, interpretable paths in the latent space of pretrained GANs, so as to provide an intuitive and easy way of controlling the underlying generative factors. In doing so, it addresses some of the limitations of the state-of-the-art works, namely, a) that they discover directions that are independent of the latent code, i.e., paths that are linear, and b) that their evaluation relies either on visual inspection or on laborious human labeling. More specifically, we propose to learn non-linear warpings on the latent space, each one parametrized by a set of RBF-based latent space warping functions, and where each warping gives rise to a family of non-linear paths via the gradient of the function. Building on the work of Voynov and Babenko, that discovers linear paths, we optimize the trainable parameters of the set of RBFs, so as that images that are generated by codes along different paths, are easily distinguishable by a discriminator network. This leads to easily distinguishable image transformations, such as pose and facial expressions in facial images. We show that linear paths can be derived as a special case of our method, and show experimentally that non-linear paths in the latent space lead to steeper, more disentangled and interpretable changes in the image space than in state-of-the art methods, both qualitatively and quantitatively. We make the code and the pretrained models publicly available at: https://github.com/chi0tzp/WarpedGANSpace.

Giorgos Kordopatis-Zilos, Christos Tzelepis, Symeon Papadopoulos et al.

In this paper, we address the problem of high performance and computationally efficient content-based video retrieval in large-scale datasets. Current methods typically propose either: (i) fine-grained approaches employing spatio-temporal representations and similarity calculations, achieving high performance at a high computational cost or (ii) coarse-grained approaches representing/indexing videos as global vectors, where the spatio-temporal structure is lost, providing low performance but also having low computational cost. In this work, we propose a Knowledge Distillation framework, called Distill-and-Select (DnS), that starting from a well-performing fine-grained Teacher Network learns: a) Student Networks at different retrieval performance and computational efficiency trade-offs and b) a Selector Network that at test time rapidly directs samples to the appropriate student to maintain both high retrieval performance and high computational efficiency. We train several students with different architectures and arrive at different trade-offs of performance and efficiency, i.e., speed and storage requirements, including fine-grained students that store/index videos using binary representations. Importantly, the proposed scheme allows Knowledge Distillation in large, unlabelled datasets -- this leads to good students. We evaluate DnS on five public datasets on three different video retrieval tasks and demonstrate a) that our students achieve state-of-the-art performance in several cases and b) that the DnS framework provides an excellent trade-off between retrieval performance, computational speed, and storage space. In specific configurations, the proposed method achieves similar mAP with the teacher but is 20 times faster and requires 240 times less storage space. The collected dataset and implementation are publicly available: https://github.com/mever-team/distill-and-select.

Zheng Gao, Ioannis Patras

Self-supervised pre-training has been proved to be effective in learning transferable representations that benefit various visual tasks. This paper asks this question: can self-supervised pre-training learn general facial representations for various facial analysis tasks? Recent efforts toward this goal are limited to treating each face image as a whole, i.e., learning consistent facial representations at the image-level, which overlooks the consistency of local facial representations (i.e., facial regions like eyes, nose, etc). In this work, we make a first attempt to propose a novel self-supervised facial representation learning framework to learn consistent global and local facial representations, Facial Region Awareness (FRA). Specifically, we explicitly enforce the consistency of facial regions by matching the local facial representations across views, which are extracted with learned heatmaps highlighting the facial regions. Inspired by the mask prediction in supervised semantic segmentation, we obtain the heatmaps via cosine similarity between the per-pixel projection of feature maps and facial mask embeddings computed from learnable positional embeddings, which leverage the attention mechanism to globally look up the facial image for facial regions. To learn such heatmaps, we formulate the learning of facial mask embeddings as a deep clustering problem by assigning the pixel features from the feature maps to them. The transfer learning results on facial classification and regression tasks show that our FRA outperforms previous pre-trained models and more importantly, using ResNet as the unified backbone for various tasks, our FRA achieves comparable or even better performance compared with SOTA methods in facial analysis tasks.

Zhonglin Sun, Chen Feng, Ioannis Patras et al.

In this work we focus on learning facial representations that can be adapted to train effective face recognition models, particularly in the absence of labels. Firstly, compared with existing labelled face datasets, a vastly larger magnitude of unlabeled faces exists in the real world. We explore the learning strategy of these unlabeled facial images through self-supervised pretraining to transfer generalized face recognition performance. Moreover, motivated by one recent finding, that is, the face saliency area is critical for face recognition, in contrast to utilizing random cropped blocks of images for constructing augmentations in pretraining, we utilize patches localized by extracted facial landmarks. This enables our method - namely LAndmark-based Facial Self-supervised learning LAFS), to learn key representation that is more critical for face recognition. We also incorporate two landmark-specific augmentations which introduce more diversity of landmark information to further regularize the learning. With learned landmark-based facial representations, we further adapt the representation for face recognition with regularization mitigating variations in landmark positions. Our method achieves significant improvement over the state-of-the-art on multiple face recognition benchmarks, especially on more challenging few-shot scenarios.

Stella Bounareli, Christos Tzelepis, Vasileios Argyriou et al.

In this paper, we present our framework for neural face/head reenactment whose goal is to transfer the 3D head orientation and expression of a target face to a source face. Previous methods focus on learning embedding networks for identity and head pose/expression disentanglement which proves to be a rather hard task, degrading the quality of the generated images. We take a different approach, bypassing the training of such networks, by using (fine-tuned) pre-trained GANs which have been shown capable of producing high-quality facial images. Because GANs are characterized by weak controllability, the core of our approach is a method to discover which directions in latent GAN space are responsible for controlling head pose and expression variations. We present a simple pipeline to learn such directions with the aid of a 3D shape model which, by construction, inherently captures disentangled directions for head pose, identity, and expression. Moreover, we show that by embedding real images in the GAN latent space, our method can be successfully used for the reenactment of real-world faces. Our method features several favorable properties including using a single source image (one-shot) and enabling cross-person reenactment. Extensive qualitative and quantitative results show that our approach typically produces reenacted faces of notably higher quality than those produced by state-of-the-art methods for the standard benchmarks of VoxCeleb1 & 2.

Stella Bounareli, Christos Tzelepis, Vasileios Argyriou et al.

Video-driven neural face reenactment aims to synthesize realistic facial images that successfully preserve the identity and appearance of a source face, while transferring the target head pose and facial expressions. Existing GAN-based methods suffer from either distortions and visual artifacts or poor reconstruction quality, i.e., the background and several important appearance details, such as hair style/color, glasses and accessories, are not faithfully reconstructed. Recent advances in Diffusion Probabilistic Models (DPMs) enable the generation of high-quality realistic images. To this end, in this paper we present DiffusionAct, a novel method that leverages the photo-realistic image generation of diffusion models to perform neural face reenactment. Specifically, we propose to control the semantic space of a Diffusion Autoencoder (DiffAE), in order to edit the facial pose of the input images, defined as the head pose orientation and the facial expressions. Our method allows one-shot, self, and cross-subject reenactment, without requiring subject-specific fine-tuning. We compare against state-of-the-art GAN-, StyleGAN2-, and diffusion-based methods, showing better or on-par reenactment performance.

Omnia Alwazzan, Abbas Khan, Ioannis Patras et al.

Brain tumors are an abnormal growth of cells in the brain. They can be classified into distinct grades based on their growth. Often grading is performed based on a histological image and is one of the most significant predictors of a patients prognosis, the higher the grade, the more aggressive the tumor. Correct diagnosis of a tumor grade remains challenging. Though histopathological grading has been shown to be prognostic, results are subject to interobserver variability, even among experienced pathologists. Recently, the World Health Organization reported that advances in molecular genetics have led to improvements in tumor classification. This paper seeks to integrate histological images and genetic data for improved computer-aided diagnosis. We propose a novel Multi-modal Outer Arithmetic Block (MOAB) based on arithmetic operations to combine latent representations of the different modalities for predicting the tumor grade (Grade \rom{2}, \rom{3} and \rom{4}). Extensive experiments evaluate the effectiveness of our approach. By applying MOAB to The Cancer Genome Atlas (TCGA) glioma dataset, we show that it can improve separation between similar classes (Grade \rom{2} and \rom{3}) and outperform prior state-of-the-art grade classification techniques.

Abhishek Kumar Singh, Ioannis Patras

The rapid evolution of the fashion industry increasingly intersects with technological advancements, particularly through the integration of generative AI. This study introduces a novel generative pipeline designed to transform the fashion design process by employing latent diffusion models. Utilizing ControlNet and LoRA fine-tuning, our approach generates high-quality images from multimodal inputs such as text and sketches. We leverage and enhance state-of-the-art virtual try-on datasets, including Multimodal Dress Code and VITON-HD, by integrating sketch data. Our evaluation, utilizing metrics like FID, CLIP Score, and KID, demonstrates that our model significantly outperforms traditional stable diffusion models. The results not only highlight the effectiveness of our model in generating fashion-appropriate outputs but also underscore the potential of diffusion models in revolutionizing fashion design workflows. This research paves the way for more interactive, personalized, and technologically enriched methodologies in fashion design and representation, bridging the gap between creative vision and practical application.

Anxhelo Diko, Tinghuai Wang, Wassim Swaileh et al.

Vision-Language Models (VLMs) are crucial for applications requiring integrated understanding textual and visual information. However, existing VLMs struggle with long videos due to computational inefficiency, memory limitations, and difficulties in maintaining coherent understanding across extended sequences. To address these challenges, we introduce ReWind, a novel memory-based VLM designed for efficient long video understanding while preserving temporal fidelity. ReWind operates in a two-stage framework. In the first stage, ReWind maintains a dynamic learnable memory module with a novel \textbf{read-perceive-write} cycle that stores and updates instruction-relevant visual information as the video unfolds. This module utilizes learnable queries and cross-attentions between memory contents and the input stream, ensuring low memory requirements by scaling linearly with the number of tokens. In the second stage, we propose an adaptive frame selection mechanism guided by the memory content to identify instruction-relevant key moments. It enriches the memory representations with detailed spatial information by selecting a few high-resolution frames, which are then combined with the memory contents and fed into a Large Language Model (LLM) to generate the final answer. We empirically demonstrate ReWind's superior performance in visual question answering (VQA) and temporal grounding tasks, surpassing previous methods on long video benchmarks. Notably, ReWind achieves a +13\% score gain and a +12\% accuracy improvement on the MovieChat-1K VQA dataset and an +8\% mIoU increase on Charades-STA for temporal grounding.

Chen Feng, Zhuo Zhi, Zhao Huang et al.

Statistically consistent methods based on the noise transition matrix ($T$) offer a theoretically grounded solution to Learning with Noisy Labels (LNL), with guarantees of convergence to the optimal clean-data classifier. In practice, however, these methods are often outperformed by empirical approaches such as sample selection, and this gap is usually attributed to the difficulty of accurately estimating $T$. The common assumption is that, given a perfect $T$, noise-correction methods would recover their theoretical advantage. In this work, we put this longstanding hypothesis to a decisive test. We conduct experiments under idealized conditions, providing correction methods with a perfect, oracle transition matrix. Even under these ideal conditions, we observe that these methods still suffer from performance collapse during training. This compellingly demonstrates that the failure is not fundamentally a $T$-estimation problem, but stems from a more deeply rooted flaw. To explain this behaviour, we provide a unified analysis that links three levels: macroscopic convergence states, microscopic optimisation dynamics, and information-theoretic limits on what can be learned from noisy labels. Together, these results give a formal account of why ideal noise correction fails and offer concrete guidance for designing more reliable methods for learning with noisy labels.

Yu Cao, Zengqun Zhao, Ioannis Patras et al.

Visual artifacts remain a persistent challenge in diffusion models, even with training on massive datasets. Current solutions primarily rely on supervised detectors, yet lack understanding of why these artifacts occur in the first place. In our analysis, we identify three distinct phases in the diffusion generative process: Profiling, Mutation, and Refinement. Artifacts typically emerge during the Mutation phase, where certain regions exhibit anomalous score dynamics over time, causing abrupt disruptions in the normal evolution pattern. This temporal nature explains why existing methods focusing only on spatial uncertainty of the final output fail at effective artifact localization. Based on these insights, we propose ASCED (Abnormal Score Correction for Enhancing Diffusion), that detects artifacts by monitoring abnormal score dynamics during the diffusion process, with a trajectory-aware on-the-fly mitigation strategy that appropriate generation of noise in the detected areas. Unlike most existing methods that apply post hoc corrections, \eg, by applying a noising-denoising scheme after generation, our mitigation strategy operates seamlessly within the existing diffusion process. Extensive experiments demonstrate that our proposed approach effectively reduces artifacts across diverse domains, matching or surpassing existing supervised methods without additional training.

Zahraa Al Sahili, Ioannis Patras, Matthew Purver

Vision-language models (VLMs) deliver strong zero-shot recognition but frequently inherit social biases from their training data. We systematically disentangle three design factors -- model size, training-data scale, and training-data source -- by comparing CLIP and OpenCLIP, two models that share an identical contrastive objective yet differ in encoder width and in the image-text corpora on which they are pre-trained (400M proprietary pairs vs. 400M/2B LAION). Across balanced face-analysis benchmarks, enlarging the encoder reduces gender skew in CLIP but amplifies both gender and racial skew in OpenCLIP; increasing the LAION corpus from 400M to 2B further increases OpenCLIP bias. At matched model and data budgets, substituting proprietary data with LAION improves gender fairness while increasing racial skew, underscoring data source as the primary driver of bias patterns. We also evaluate three post-hoc, test-time debiasing strategies -- Bias Prompts, Prompt Array, and SANER. Debiasing reduces but does not eliminate harm, and its effectiveness is source- and size-dependent: Bias Prompts most effectively reduce gender skew in CLIP at smaller model sizes, whereas Prompt Array and SANER more reliably reduce racial skew in OpenCLIP; scaling LAION reconfigures which method is most fair. Taken together, these findings challenge the assumption that bigger models or datasets are automatically fairer and foreground training data source as the key determinant of both bias and mitigation efficacy. We release code and evaluation scripts to enable transparent, reproducible auditing of future VLMs.

Zahraa Al Sahili, Ioannis Patras, Matthew Purver

Multilingual vision-language models (VLMs) promise universal image-text retrieval, yet their social biases remain underexplored. We perform the first systematic audit of four public multilingual CLIP variants: M-CLIP, NLLB-CLIP, CAPIVARA-CLIP, and the debiased SigLIP-2, covering ten languages that differ in resource availability and morphological gender marking. Using balanced subsets of FairFace and the PATA stereotype suite in a zero-shot setting, we quantify race and gender bias and measure stereotype amplification. Contrary to the intuition that multilinguality mitigates bias, every model exhibits stronger gender skew than its English-only baseline. CAPIVARA-CLIP shows its largest biases precisely in the low-resource languages it targets, while the shared encoder of NLLB-CLIP and SigLIP-2 transfers English gender stereotypes into gender-neutral languages; loosely coupled encoders largely avoid this leakage. Although SigLIP-2 reduces agency and communion skews, it inherits -- and in caption-sparse contexts (e.g., Xhosa) amplifies -- the English anchor's crime associations. Highly gendered languages consistently magnify all bias types, yet gender-neutral languages remain vulnerable whenever cross-lingual weight sharing imports foreign stereotypes. Aggregated metrics thus mask language-specific hot spots, underscoring the need for fine-grained, language-aware bias evaluation in future multilingual VLM research.

Mariano V. Ntrougkas, Vasileios Mezaris, Ioannis Patras

The adoption of Deep Neural Networks (DNNs) in critical fields where predictions need to be accompanied by justifications is hindered by their inherent black-box nature. In this paper, we introduce P-TAME (Perturbation-based Trainable Attention Mechanism for Explanations), a model-agnostic method for explaining DNN-based image classifiers. P-TAME employs an auxiliary image classifier to extract features from the input image, bypassing the need to tailor the explanation method to the internal architecture of the backbone classifier being explained. Unlike traditional perturbation-based methods, which have high computational requirements, P-TAME offers an efficient alternative by generating high-resolution explanations in a single forward pass during inference. We apply P-TAME to explain the decisions of VGG-16, ResNet-50, and ViT-B-16, three distinct and widely used image classifiers. Quantitative and qualitative results show that our method matches or outperforms previous explainability methods, including model-specific approaches. Code and trained models will be released upon acceptance.