Recent advances in interactive text-to-image retrieval (I-TIR) use diffusion models to bridge the modality gap between the textual information need and the images to be searched. However, existing frameworks fuse multi-modal user feedback by simple embedding addition. In this work, we show that this basic fusion strategy indiscriminately incorporates generative noise produced by the diffusion model, leading to performance degradation for up to 55.62% of samples. We further propose ADaFuSE (Adaptive Diffusion-Text Fusion with Semantic-aware Experts), a lightweight fusion model designed to align and calibrate multi-modal views for diffusion-augmented I-TIR, which can be plugged into existing frameworks without modifying the backbone encoder. Specifically, we introduce a dual-branch fusion mechanism that employs an adaptive gating branch to dynamically balance modality reliability, alongside a semantic-aware mixture-of-experts branch to capture fine-grained cross-modal nuances. Via thorough evaluation over four standard I-TIR benchmarks, ADaFuSE achieves state-of-the-art performance, surpassing the DAR model by up to 3.49% in Hits@10 with only a 5.29% parameter increase, while exhibiting stronger robustness to noisy and longer interactive queries. These results show that generative augmentation coupled with principled fusion provides a simple, generalizable alternative to fine-tuning for tasks like product search.

Interactive Text-to-image retrieval (I-TIR) is an important enabler for a wide range of state-of-the-art services in domains such as e-commerce and education.However, current methods rely on finetuned Multimodal Large Language Models (MLLMs), which are costly to train and update, and exhibit poor generalizability.This latter issue is of particular concern, as: 1) finetuning narrows the pretrained distribution of MLLMs, thereby reducing generalizability; and 2) I-TIR introduces increasing query diversity and complexity.As a result, I-TIR solutions are highly likely to encounter queries and images not well represented in any training dataset.To address this, we propose leveraging Diffusion Models (DMs) for text-to-image mapping, to avoid finetuning MLLMs while preserving robust performance on complex queries.Specifically, we introduce Diffusion Augmented Retrieval (DAR), a framework that generates multiple intermediate representations via LLM-based dialogue refinements and DMs, producing a richer depiction of the user's information needs.This augmented representation facilitates more accurate identification of semantically and visually related images.Extensive experiments on four benchmarks show that for simple queries, DAR achieves results on par with finetuned I-TIR models, yet without incurring their tuning overhead.Moreover, as queries become more complex through additional conversational turns, DAR surpasses finetuned I-TIR models by up to 7.61% in Hits@10 after ten turns, illustrating its improved generalization for more intricate queries.

In recent years, the rapid growth of online multimedia services, such as e-commerce platforms, has necessitated the development of personalised recommendation approaches that can encode diverse content about each item. Indeed, modern multi-modal recommender systems exploit diverse features obtained from raw images and item descriptions to enhance the recommendation performance. However, the existing multi-modal recommender systems primarily depend on the features extracted individually from different media through pre-trained modality-specific encoders, and exhibit only shallow alignments between different modalities, thereby limiting these systems’ ability to capture the underlying relationships between the modalities. In this article, we enhance the deep alignment of large multi-modal encoders to address the shallow alignment of modalities in multi-modal recommender systems. These encoders have previously demonstrated state-of-the-art effectiveness in ranking items across various domains. Specifically, we investigate the use of three state-of-the-art large multi-modal encoders – CLIP (dual-stream), VLMo and BEiT-3 (unified) – for recommendation tasks. We explore their benefits for recommendation through using a range of strategies, including the use of pre-trained and fine-tuned encoders, as well as the evaluation of the end-to-end training of these encoders. We show that pre-trained large multi-modal encoders generate more aligned and effective user/item representations compared with existing modality-specific encoders across four existing multi-modal recommendation datasets. Furthermore, we show that fine-tuning these encoders further improves the recommendation performance, with end-to-end training emerging as the most effective paradigm, significantly outperforming both pre-trained and fine-tuned encoders with an improved recommendation performance. We also demonstrate the effectiveness of large multi-modal encoders in facilitating modality alignment by evaluating the contribution of each modality separately. Finally, we show that the dual-stream approach, specifically CLIP, is the most effective architecture for these large multi-modal encoders, outperforming the unified approaches (i.e., VLMo and BEiT3) in terms of effectiveness and efficiency.

State-of-the-art pre-trained image models predominantly adopt a two-stage approach: initial unsupervised pre-training on large-scale datasets followed by task-specific fine-tuning using Cross-Entropy loss (CE). However, it has been demonstrated that CE can compromise model generalization and stability. While recent works employing contrastive learning address some of these limitations by enhancing the quality of embeddings and producing better decision boundaries, they often overlook the importance of hard negative mining and rely on resource intensive and slow training using large sample batches. To counter these issues, we introduce a novel approach named CLCE, which integrates Label-Aware Contrastive Learning with CE. Our approach not only maintains the strengths of both loss functions but also leverages hard negative mining in a synergistic way to enhance performance. Experimental results demonstrate that CLCE significantly outperforms CE in Top-1 accuracy across twelve benchmarks, achieving gains of up to 3.52% in few-shot learning scenarios and 3.41% in transfer learning settings with the BEiT-3 model. Importantly, our proposed CLCE approach effectively mitigates the dependency of contrastive learning on large batch sizes such as 4096 samples per batch, a limitation that has previously constrained the application of contrastive learning in budget-limited hardware environments.

As Multimodal Large Language Models (MLLMs) grow in size, adapting them to specialized tasks becomes increasingly challenging due to high computational and memory demands. Indeed, traditional fine-tuning methods are costly, due to the need for extensive, task-specific training. While efficient adaptation methods exist that aim to reduce these costs, in practice they suffer from shallow inter-modal alignment, which severely hurts model effectiveness. To tackle these computational challenges and improve inter-modal alignment, we introduce the MultiWay-Adapter (MWA), a novel framework featuring an ‘Alignment Enhancer’. This enhancer deepens inter-modal alignment, enabling high transferability with minimal tuning effort. Our experiments show that unlike prior efficient tuning approaches, MWA maintains model effectiveness, while reducing training time by up-to 57%. MWA is also lightweight, increasing model size by only 2-3% (in terms of parameters) for state-of-the-art foundation models like BEiT-3 Large. These results demonstrate that MWA provides an efficient and effective adaptation method for MLLMs, significantly broadening their applicability.

Vision Transformers (ViTs) have emerged as popular models in computer vision, demonstrating state-of-the-art performance across various tasks. This success typically follows a two-stage strategy involving pre-training on large-scale datasets using self-supervised signals, such as masked random patches, followed by fine-tuning on task-specific labeled datasets with cross-entropy loss. However, this reliance on cross-entropy loss has been identified as a limiting factor in ViTs, affecting their generalization and transferability to downstream tasks. Addressing this critical challenge, we introduce a novel Label-aware Contrastive Training framework, LaCViT, which significantly enhances the quality of embeddings in ViTs. LaCViT not only addresses the limitations of cross-entropy loss but also facilitates more effective transfer learning across diverse image classification tasks. Our comprehensive experiments on eight standard image classification datasets reveal that LaCViT statistically significantly enhances the performance of three evaluated ViTs by up-to 10.78% under Top-1 Accuracy.

Robotic vision applications often necessitate a wide range of visual perception tasks, such as object detection, segmentation, and identification. While there have been substantial advances in these individual tasks, integrating specialized models into a unified vision pipeline presents significant engineering challenges and costs. Recently, Multimodal Large Language Models (MLLMs) have emerged as novel backbones for various downstream tasks. We argue that leveraging the pre-training capabilities of MLLMs enables the creation of a simplified framework, thus mitigating the need for task-specific encoders. Specifically, the large-scale pretrained knowledge in MLLMs allows for easier fine-tuning to downstream robotic vision tasks and yields superior performance. We introduce the RoboLLM framework, equipped with a BEiT-3 backbone, to address all visual perception tasks in the ARMBench challenge—a large-scale robotic manipulation dataset about real-world warehouse scenarios. RoboLLM not only outperforms existing baselines but also substantially reduces the engineering burden associated with model selection and tuning. All the code used in this paper can be found in https://github.com/longkukuhi/RoboLLM.

Text-to-image retrieval aims to find the relevant images based on a text query, which is important in various use-cases, such as digital libraries, e-commerce, and multimedia databases. Although Multimodal Large Language Models (MLLMs) demonstrate state-of-the-art performance, they exhibit limitations in handling large-scale, diverse, and ambiguous real-world needs of retrieval, due to the computation cost and the injective embeddings they produce. This paper presents a two-stage Coarse-to-Fine Index-shared Retrieval (CFIR) framework, designed for fast and effective large-scale long-text to image retrieval. The first stage, Entity-based Ranking (ER), adapts to long-text query ambiguity by employing a multiple-queries-to-multiple-targets paradigm, facilitating candidate filtering for the next stage. The second stage, Summary-based Re-ranking (SR), refines these rankings using summarized queries. We also propose a specialized Decoupling-BEiT-3 encoder, optimized for handling ambiguous user needs and both stages, which also enhances computational efficiency through vector-based similarity inference. Evaluation on the AToMiC dataset reveals that CFIR surpasses existing MLLMs by up to 11.06% in Recall@1000, while reducing training and retrieval times by 68.75% and 99.79%, respectively. We will release our code to facilitate future research at https://github.com/longkukuhi/CFIR.

In many-core chips seen in mobile computing, data center, AI, and elsewhere, thermal covert channels could be established to transmit data (e.g., passwords), supposedly to be kept secret and private. Effectiveness of a thermal covert channel, measured by its transmission rate and bit error rate (BER), is so much dependent on the thermal noise/interference imposed on the channel. In this paper, we present a few techniques to improve the capacity of thermal covert channel by overcoming the thermal interference. In particular, data in a thermal covert channel are encoded and represented following a new thermal signaling scheme where logic value, 0 or 1, modules the thermal signals duty cycle. Next, we show in this study that proper selection of transmission frequency can significantly minimize thermal interference. In addition, we propose a robust end-to-end communication protocol for reliable communications. Our experiments have confirmed that, compared to an existing thermal covert channel attack [1] [2], a thermal covert channel enhanced with all the improvements proposed in this study is seeing significant BER reduction (by as much as 75%), and transmission rate boost (by more than threefold). Building such a strong thermal covert channel is the key step towards developing robust defense and countermeasures against information leaking over thermal covert channel.