TAME: Temporal-Aware Mixture-of-Experts for Text-Video Retrieval

Text-Video Retrieval (TVR) retrieves videos that match a natural-language query, but extending image-text models such as CLIP to videos is fundamentally limited by the lack of temporal modeling. Videos exhibit frame-wise heterogeneity in appearance and motion, and compressing all frames into a single representation often obscures temporal structure and semantic transitions. To address this, we propose Temporal-Aware Mixture-of-Experts for Text-Video Retrieval (TAME), a CLIP-based framework that jointly models frame-level structure and temporal relations. First, we integrate sparse Mixture-of-Experts (MoE) layers into both CLIP encoders and apply frame-consistent routing on the vision branch so that experts specialize according to frame-level visual patterns while preserving the original vision-language alignment. Second, we introduce Frame-Temporal (FT) tokens that aggregate global cross-frame information and feed it back to each frame, enabling the visual encoder to capture long-range temporal dependencies without harming local details. Third, we design a Cross-Temporal Interaction and Aggregation (CTIA) module that refines frame-wise sentence-video similarities through staged temporal filtering and fusion. Experiments on standard TVR benchmarks show that TAME consistently improves over CLIP-based baselines. On MSR-VTT, it improves R@1 by 4.0 over CLIP4Clip, and also achieves consistent gains on DiDeMo, MSVD, LSMDC, and ActivityNet. The code is available at https://github.com/sejong-rcv/TAME.

ZOMP: Zeroth-Order Multi-Modal Prompt Tuning for Vision-Language Models

Fine-tuning vision-language models such as CLIP typically requires backpropagation (BP) through the full model, which is infeasible when only forward-pass access is available, as is common for memory-constrained edge devices and proprietary model deployments. Prior BP-free, zeroth-order prompt-tuning methods avoid this requirement but often tune prompts in a single modality or optimize over a search space large enough that convergence requires thousands of forward passes, which is impractical under realistic query budgets. We propose ZOMP (Zeroth-Order Multimodal Prompt tuning), a query-efficient, fully forward-only method that tunes deep prompts in both the vision and text branches of a frozen CLIP model using simultaneous perturbation stochastic approximation. ZOMP combines three ingredients: a cross-modal low-rank reparameterization that ties the two branches through a shared factor and keeps the effective search dimensionality small, a gradient-correction momentum term that stabilizes the noisy zeroth-order estimate, and a budget-indexed rank schedule that unlocks capacity as the query budget is spent. Across 13 vision-language benchmarks under a matched 5,000-query budget, ZOMP consistently outperforms prior BP-free prompt-tuning methods in both few-shot accuracy and query efficiency, and it generalizes better across base-to-new, cross-dataset transfer, and out-of-distribution settings. Our results show that jointly exploiting multimodality and low-rank structure is an effective route to practical, query-efficient BP-free prompt tuning.

Controlling Embedding Spaces with Text-Conditioned Transformations

Multimodal embedding spaces in models like CLIP enable powerful capabilities such as semantic similarity retrieval and cross-modal zero-shot classification. These embeddings compress high-level semantics into a single vector, which comes at the cost of primarily expressing a dominant semantics like main object while suppressing other important attributes such as camera angle or color tone. We propose a text-conditioned transformation of visual embeddings that makes such attributes explicitly accessible. Given a natural language description of an attribute category (e.g., "color" or "art style"), a network generates an affine transformation that emphasizes the specified attribute. Conditioning on text enables it to learn many attributes simultaneously, accessing them at inference time through an intuitive interface. The network is trained to align transformed embeddings with the frozen latent space, enabling retrieval using existing large-scale embeddings without any re-encoding. When applied to a full set, the same mechanism transforms the latent space for attribute disentanglement tasks such as multi-clustering. By operating directly in latent space, our method provides a unified and efficient framework for controlling embedding spaces, demonstrating state-of-the-art performance across both attribute-based retrieval and multi-attribute organization tasks with near-zero inference cost. Project page: https://joefioresi718.github.io/ControlEmbed_webpage/

Now You See the Hate: Adaptive View Retrieval for Hidden Hateful Illusions

Hateful optical illusions expose a serious gap in current multimodal safety systems. On original-view hateful illusions, previous work shows that six moderation classifiers achieve at most 20.9 to 24.5% accuracy and nine state-of-the-art VLMs remain at or below 10.2% with illusion-aware prompting, leaving most hidden hate undetected. We formulate hidden hateful illusion detection as a perceptual retrieval problem and propose Adaptive View Retrieval. This retrieve-and-calibrate framework assembles a complementary view bank for the image and hidden-message templates, adaptively selects which views to trust, retrieves hidden-message identities, and calibrates whether the recovered evidence is harmful. On HatefulIllusion with a frozen CLIP encoder, Adaptive View Retrieval reaches 93.2% balanced accuracy on the held-out test split. It substantially outperforms original-view baselines and fixed single-transform filters across hate slangs, hate symbols, and visibility levels. The same design also surpasses official fine-tuned CLIP baselines, matches or exceeds human performance on IllusionMNIST, IllusionFashionMNIST, and IllusionAnimals, and outperforms zoom-out preprocessing on HC-Bench under the SemVink protocol. Together, these results show that robust multimodal moderation requires recovering hidden meaning before deciding whether it is harmful.

AlphaWiSE: Adaptive Weight Interpolation for Continual Multimodal Representation Learning

Multimodal models such as CLIP learn a shared embedding space for cross-modal retrieval, but continual adaptation to sequentially arriving data can disrupt the cross-modal alignment acquired from earlier phases. Conventional continual-learning methods return a single checkpoint, which commits every retrieval direction to the same stability-plasticity trade-off. We propose AlphaWiSE, a post-hoc weight-space interpolation method that composes two frozen source checkpoints. For each aligned parameter tensor identified by its checkpoint key, AlphaWiSE fits one scalar interpolation coefficient shared by all tensor entries. The coefficients are fitted on a smaller exemplar memory and used to materialize one interpolated checkpoint. The deployed model has the same architecture and parameter count as either source checkpoint, which does not require additional inference time. Extensive experiments on audio-image-text retrieval show consistent improvements over strong continual-learning baselines across multiple retrieval directions and evaluation metrics.

The Hyperspherical Geometry of CLIP Latent Space: A Semantic Mixture Model

Contrastive Language-Image Pretraining (CLIP) representations form a semantic embedding space governed by cosine similarity, reflecting an intrinsic hyperspherical geometry. However, existing probabilistic interpretations typically rely on Gaussian assumptions, which fail to capture this directional and multimodal structure. We propose a principled density model for the CLIP latent space based on Mixtures of von Mises-Fisher (MovMF) distributions defined on the unit hypersphere. Using the Expectation-Maximization (EM) algorithm, we efficiently learn a probabilistic model in which each mixture component corresponds to a coherent semantic concept. This formulation yields a closed-form likelihood naturally aligned with hyperspherical geometry, enabling accurate and interpretable density estimation. Empirically, our model significantly improves long-tailed and out-of-distribution detection and provides a natural semantic decomposition, representing each embedding as a sparse probabilistic combination of interpretable concepts. These results suggest that CLIP latent space is more faithfully characterized as a hyperspherical semantic mixture rather than an isotropic Gaussian, establishing a simple and geometrically consistent probabilistic framework for modeling and understanding multimodal representations. Project page is available at https://xiaoyuzhizi.github.io/movmf-clip/.

HANCLIP: A Family of Hyperbolic Angular Negation Vision Language Models

Vision-Language Models (VLMs) are typically pre-trained on large-scale image-text datasets to capture semantic correspondences between visual content and natural language. However, they remain surprisingly brittle to negation: models often rely on shallow word co-occurrence and are easily distracted by misleading or irrelevant textual cues, even when their overall retrieval or classification performance is strong. Moreover, directly finetuning on negation data can interfere with previously acquired knowledge, causing noticeable degradation on standard vision-language benchmarks. To tackle these issues, this work introduces HANCLIP (Hyperbolic + Angular + Negation), a family of VLMs that explicitly restructures the embedding space to encode "what an image is not" alongside "what it is." HANCLIP is trained on a compact set of 20,000 image-text quadruplets and combines a hyperbolic formulation, which models hierarchical semantic relations and asymmetries, with an angular triplet objective that drives systematic separation between negated descriptions and their corresponding positives. This geometry-aware design strengthens negation sensitivity while preserving the global structure of pretrained representations, rather than overwriting them. Extensive experiments across multiple vision-language tasks show that HANCLIP delivers consistent gains on the negation-focused NegBench benchmark, while maintaining competitive or improved performance on standard classification and image-text retrieval benchmarks. The framework is model-agnostic and can be plugged into CLIP, LongCLIP, SmartCLIP, and HiMo-CLIP without large-scale retraining, demonstrating that a carefully designed geometric objective can substantially extend the reasoning capabilities of existing VLMs using only modest additional data.

Multimodal Image Colorization: Quantifying the Impact of Text-Conditioned Guidance on Grayscale-to-Color Translation

Grayscale images are commonly found in historical photography restoration, medical imaging, and artistic media. However, automatically applying color to these images remains a significant challenge in computer vision because many plausible colorizations can correspond to the same grayscale input. In this work, we quantify the effect of text conditioning on pixel-level and perceptual metrics for grayscale-to-color image models. Specifically, we compare two architectures, a U-Net and Stable Diffusion 1.5, each tested with and without CLIP text conditioning while holding all other variables constant. Our results show that text conditioning improves PSNR by 5.6%, SSIM by 1.2%, and colorfulness by 36.6%, while reducing LPIPS by 7.6% in the U-Net tier. In the Stable Diffusion tier, text conditioning improves PSNR by 5.8%, SSIM by 1.5%, and colorfulness by 0.6%, while reducing LPIPS by 11.3%. These results indicate that text conditioning provides consistent, measurable improvements to colorization quality across both architecture scales.

Cross-Modal Masked Compositional Concept Modeling for Enhancing Visio-Linguistic Compositionality

Contrastively trained vision-language models like CLIP, have made remarkable progress in learning joint image-text representations, but still face challenges in compositional understanding. They often exhibit a "bag-of-words" behavior--struggling to capture the object relations, attribute-object bindings, and word order dependencies. This limitation arises not only from the reliance on global, single-vector representations for optimization, but also from the insufficient exploitation and modeling of the rich compositional information inherently present in paired image text data. In this work, we propose MACCO (MAsked Compositional Concept MOdeling), a framework that masks compositional concepts in one modality and reconstructs them conditioned on the full contextual information from the other, enabling the model to capture and align cross-modal compositional structures more effectively. To facilitate this process, we introduce two auxiliary objectives that jointly align and regularize masked features both inter-modally and intra-modally. Extensive experiments on five compositional benchmarks, along with in-depth analyses, demonstrate that our approach not only significantly enhances compositionality in VLMs but also improves their ability to capture syntactic structure and linguistic information. Additionally, the improved compositionality also benefits text-to-image generation and multimodal large language model. Code is available at https://github.com/hiker-lw/MACCO.

Lexical discovery in unknown environments orchestrated by Large Language Models

Populations of autonomous agents deployed in unknown environments (e.g. planetary or deep-sea exploration) must develop shared vocabularies to refer to entities that have no name in any human language. We propose the Neuro-Symbolic Lexical Discovery (NSLD) framework, in which a population of LLM-based agents plays a referential game over out-of-distribution visual referents, autonomously self-organising a shared alien lexicon. Each agent combines a frozen CLIP vision encoder with a private FAISS vector index and a text-only LLM. Crucially, discovered alien words are anchored to natural language via semantic proximity in the embedding space, enlarging the human vocabulary with new perceptually grounded words. Consensus is reached in simulations with populations of up to twenty agents and ten visual referents. Convergence dynamics are characterised through three analytical models achieving R^2 > 0.95, representing a first step towards pre-deployment planning in autonomous exploration missions.

Disentangling Hallucinations: Orthogonal Semantic Projection for Robust Interpretability

As Vision-Language Models are increasingly deployed in safety-critical applications, the trustworthiness of their explanations becomes crucial. Explainable AI (XAI) methods for Vision-Language Models often suffer from semantic hallucination, where attribution maps highlight prominent image regions even when prompted with incorrect text descriptions (e.g., highlighting a dog when prompted ``cat''). Although this problem is widespread, a formal mathematical analysis of XAI methods and CLIP embeddings is largely missing in the literature. We demonstrate that this phenomenon is not specific to a single architecture but is a fundamental consequence of Linear Semantic Leakage in high-dimensional embedding spaces. We propose a unified theoretical framework, Linear Semantic Attribution (LSA), which generalizes across discriminative methods. We introduce OSP, a geometric intervention that utilizes the residual property of OMP to disentangle unique semantic signals from shared concepts. We prove theoretically and demonstrate empirically that OSP minimizes hallucination by orthogonalizing the query vector against distractor concepts, rendering the attribution model blind to shared features while preserving fidelity for correct prompts. Our code is available at: https://github.com/emirhanbilgic/Orthogonal-Semantic-Projection

TEVI: Text-Conditioned Editing of Visual Representations via Sparse Autoencoders for Improved Vision-Language Alignment

Vision-language models such as CLIP are highly useful for diverse tasks due to their shared image-text embedding space. Despite this, the image and text embeddings are often poorly aligned, affecting downstream performance. Recent work has hypothesized that this can be attributed to an information imbalance: images contain more information than their captions describe. In this work, we propose TEVI, a framework that uses captions as a signal for what to retain from image embeddings. Specifically, we use sparse autoencoders to disentangle image embeddings and train a masking module to selectively reconstruct the embedding based on a given caption. In a controlled setup with synthetic captions, we show that TEVI is effective at preserving caption-described attributes while discarding others. We find that this extends to CLIP models trained on natural images, where TEVI learns to mask meaningfully and allows retrieval based on conditioning. Finally, we use TEVI to achieve improved retrieval performance across coarse-grained and fine-grained benchmarks. Code available at https://github.com/neuroexplicit-saar/TEVI.

KODA: Contrastive Representation Comparison and Alignment for Vision-Language Foundation Models

Vision-language foundation models such as CLIP and SigLIP provide widely used representations for multimodal learning systems. While these models are typically compared through downstream performance, such evaluations often do not explain how their representations differ structurally. In this work, we study this problem through the task of Contrastive Embedding Clustering: identifying sample subsets that are weakly clustered under one representation but strongly clustered under another. We propose \emph{Kernel Optimization for Discrepancy Analysis (KODA)}, a kernel-based framework for contrastive representation comparison and alignment. KODA constructs unified multimodal kernels through modality-wise kernel composition and formulates discrepancy discovery as a constrained optimization problem that searches for coherent structures in one representation while suppressing coherence in a reference representation. This yields interpretable discrepancy directions associated with specific sample subsets and modality interactions. To scale KODA to large vision-language datasets, we develop randomized low-dimensional approximations of joint kernels using random projections, including Random Fourier Features for shift-invariant kernels. Empirically, KODA identifies consistent and interpretable discrepancy structures across vision-language representations and provides sample subsets for representation alignment. The code is available at https://github.com/yokiwuuu/KODA.