arXiv subjects

Mohammadreza Salehi

Publications and source records attributed to Mohammadreza Salehi.

At least 19 recordsLinked to original sources

arXiv2026

How Do Video Foundation Models Encode Intuitive Physics? Probing Across Pretraining Paradigms

We study whether pretrained video foundation models encode intuitive-physics information in their frozen representations, and how this information varies across model families, layers, and probe types. Using frozen-feature probing on IntPhys2 and Minimal Video Pairs (MVP), we compare predictive joint-embedding models (V-JEPA), masked reconstruction models (VideoMAE), and a diffusion-based video generator (LTX-Video). V-JEPA achieves the strongest overall results across benchmarks, especially with probes that model temporal dynamics, while VideoMAE remains competitive and LTX-Video recovers weaker but non-trivial signal. Layerwise analyses show that physics-relevant information is weakest in early layers and becomes most accessible at intermediate-to-late depth, and temporal controls show that disrupting frame order substantially reduces performance, especially on MVP. Together, these results suggest that intuitive-physics knowledge emerges reliably in pretrained video representations, but its accessibility depends strongly on pretraining paradigm, representational depth, and readout mechanism.

cs.CV↗

arXiv2026

Crane: Context-Guided Prompt Learning and Attention Refinement for Zero-Shot Anomaly Detection

Zero-shot anomaly detection and localization aims to learn from source-domain data and generalize to unseen target domains without target-domain samples. Recent CLIP-based methods perform inference by comparing visual features with normal and abnormal textual prototypes; however, dense localization remains substantially weaker than image-level detection, especially for small or subtle abnormal regions. We identify two key limitations behind this gap: CLIP's vision encoder is primarily optimized for global image-text alignment, which limits its ability to preserve fine-grained spatial details, and the learned normal/abnormal text representations are not sufficiently aligned with dense visual features. To address these limitations, we propose Crane (Context-Guided Prompt Learning and Attention Refinement), a CLIP-based framework that first adapts the vision encoder with a correlation-based attention module to better preserve local visual structure. Second, it conditions learnable normal and abnormal prompts on global image context during training, improving instance-aware text-vision alignment. Third, it incorporates anomaly-aware local-to-global fusion, which injects anomaly-relevant patch features into the global image representation for more sensitive image-level detection. Finally, we show that the same correlation-based attention design can exploit spatial correlations from DINOv2, yielding a boosted variant, Crane+, with stronger localization ability. Across seven industrial benchmarks, Crane improves mean image-level AP by 4.5% over the strongest compared baseline, while Crane+ improves mean pixel-level AUPRO by 9.0%. The code is available at https://github.com/Alireza99Salehi/Crane.

cs.CV↗

arXiv2026

A Unified Perspective on Causality and One-Sided System Responses in Time and Space Across Physical and Fourier Domains

The principle of causality has long been mathematically associated with the frameworks of Titchmarsh's theorem and the Kramers--Kronig relations. While these relations arise naturally in the context of temporal system responses -- ensuring that the effect of an applied field or force does not precede its cause -- they have recently been shown to provide a pathway for realizing one-sided system responses in a variety of physical settings. In particular, one-sided frequency responses and one-sided wavevector responses have been successfully studied and engineered, enabling the prospect of numerous applications based on the complete suppression of backward scattering. In this work, we present a brief review of causality and its connection to these Fourier-domain analogs. We then turn our attention to the only remaining setting in which a one-sided system response may be explored: one-sided spatial nonlocality. We specifically investigate the possibility of realizing a one-sided spatial response within the widely used framework of nonlocal flat optics, where we uncover fundamental obstacles that hinder the achievement of such functionality in these structures. This, in turn, raises an intriguing open question: is one-sided spatial nonlocal response merely incompatible with the specific platform of nonlocal flat optics, or is it fundamentally forbidden by nature itself?

physics.optics↗

arXiv2026

One Trajectory, One Token: Grounded Video Tokenization via Panoptic Sub-object Trajectory

Effective video tokenization is critical for scaling transformer models for long videos. Current approaches tokenize videos using space-time patches, leading to excessive tokens and computational inefficiencies. The best token reduction strategies degrade performance and barely reduce the number of tokens when the camera moves. We introduce grounded video tokenization, a paradigm that organizes tokens based on panoptic sub-object trajectories rather than fixed patches. Our method aligns with fundamental perceptual principles, ensuring that tokenization reflects scene complexity rather than video duration. We propose TrajViT, a video encoder that extracts object trajectories and converts them into semantically meaningful tokens, significantly reducing redundancy while maintaining temporal coherence. Trained with contrastive learning, TrajViT significantly outperforms space-time ViT (ViT3D) across multiple video understanding benchmarks, e.g., TrajViT outperforms ViT3D by a large margin of 6% top-5 recall in average at video-text retrieval task with 10x token deduction. We also show TrajViT as a stronger model than ViT3D for being the video encoder for modern VideoLLM, obtaining an average of 5.2% performance improvement across 6 VideoQA benchmarks while having 4x faster training time and 18x less inference FLOPs. TrajViT is the first efficient encoder to consistently outperform ViT3D across diverse video analysis tasks, making it a robust and scalable solution.

cs.CV↗

arXiv2026

VideoNet: A Large-Scale Dataset for Domain-Specific Action Recognition

Videos are unique in their ability to capture actions which transcend multiple frames. Accordingly, for many years action recognition was the quintessential task for video understanding. Unfortunately, due to a lack of sufficiently diverse and challenging data, modern vision-language models (VLMs) are no longer evaluated on their action recognition capabilities. To revitalize action recognition in the era of VLMs, we advocate for a returned focus on domain-specific actions. To this end, we introduce VideoNet, a domain-specific action recognition benchmark covering 1,000 distinct actions from 37 domains. We begin with a multiple-choice evaluation setting, where the difference between closed and open models is stark: Gemini 3.1 Pro attains 69.9% accuracy while Qwen3-VL-8B gets a mere 45.0%. To understand why VLMs struggle on VideoNet, we relax the questions into a binary setting, where random chance is 50%. Still, Qwen achieves only 59.2% accuracy. Further relaxing the evaluation setup, we provide $k\in\{1,2,3\}$ in-context examples of the action. Some models excel in the few-shot setting, while others falter; Qwen improves $+7.0\%$, while Gemini declines $-4.8\%$. Notably, these gains fall short of the $+13.6\%$ improvement in non-expert humans when given few-shot examples. Finding that VLMs struggle to fully exploit in-context examples, we shift from test-time improvements to the training side. We collect the first large-scale training dataset for domain-specific actions, totaling nearly 500k video question-answer pairs. Fine-tuning a Molmo2-4B model on our data, we surpass all open-weight 8B models on the VideoNet benchmark.

cs.CV↗

arXiv2026

Franca: Nested Matryoshka Clustering for Scalable Visual Representation Learning

We present Franca (pronounced Fran-ka): free one; the first fully open-source (data, code, weights) vision foundation model that matches and in many cases surpasses the performance of state-of-the-art proprietary models, e.g., DINOv2, CLIP, SigLIPv2, etc. Our approach is grounded in a transparent training pipeline inspired by Web-SSL and uses publicly available data: ImageNet-21K and a subset of ReLAION-2B. Beyond model release, we tackle critical limitations in SSL clustering methods. While modern models rely on assigning image features to large codebooks via clustering algorithms like Sinkhorn-Knopp, they fail to account for the inherent ambiguity in clustering semantics. To address this, we introduce a parameter-efficient, multi-head clustering projector based on nested Matryoshka representations. This design progressively refines features into increasingly fine-grained clusters without increasing the model size, enabling both performance and memory efficiency. Additionally, we propose a novel positional disentanglement strategy that explicitly removes positional biases from dense representations, thereby improving the encoding of semantic content. This leads to consistent gains on several downstream benchmarks, demonstrating the utility of cleaner feature spaces. Our contributions establish a new standard for transparent, high-performance vision models and open a path toward more reproducible and generalizable foundation models for the broader AI community. The code and model checkpoints are available at https://github.com/valeoai/Franca.

cs.CV↗

arXiv2026

Nonlocal photonic time crystals: Infinite momentum bandgaps with minimal modulation speed and strength

For over a decade, photonic time crystals have promised access to novel and exotic optical phenomena, offering fundamentally new ways to manipulate classical and quantum light. Central to these capabilities is the emergence of momentum bandgaps -- the counterpart of the more familiar frequency bandgaps in spatial crystals -- which have proven difficult to observe experimentally due to the combined need for high modulation speed and strength. To date, these requirements have all but hindered the development of time crystals at optical frequencies. Here, we show that the stringent modulation-speed requirement is a direct consequence of the Manley-Rowe relations governing conventional modulation schemes. We further demonstrate that modulating the plasma frequency of a Lorentz-dispersive material overcomes this limitation. Incorporating a specific form of spatial nonlocality (spatial dispersion) into this already temporally nonlocal (frequency dispersive) framework removes all remaining constraints, enabling momentum bandgaps of infinite extent -- in both frequency and momentum -- with arbitrarily small modulation speeds and strengths.

physics.optics↗

arXiv2026

Molmo2: Open Weights and Data for Vision-Language Models with Video Understanding and Grounding

Today's strongest video-language models (VLMs) remain proprietary. The strongest open-weight models either rely on synthetic data from proprietary VLMs, effectively distilling from them, or do not disclose their training data or recipe. As a result, the open-source community lacks the foundations needed to improve on the state-of-the-art video (and image) language models. Crucially, many downstream applications require more than just high-level video understanding; they require grounding -- either by pointing or by tracking in pixels. Even proprietary models lack this capability. We present Molmo2, a new family of VLMs that are state-of-the-art among open-source models and demonstrate exceptional new capabilities in point-driven grounding in single image, multi-image, and video tasks. Our key contribution is a collection of 7 new video datasets and 2 multi-image datasets, including a dataset of highly detailed video captions for pre-training, a free-form video Q&A dataset for fine-tuning, a new object tracking dataset with complex queries, and an innovative new video pointing dataset, all collected without the use of closed VLMs. We also present a training recipe for this data utilizing an efficient packing and message-tree encoding scheme, and show bi-directional attention on vision tokens and a novel token-weight strategy improves performance. Our best-in-class 8B model outperforms others in the class of open weight and data models on short videos, counting, and captioning, and is competitive on long-videos. On video-grounding Molmo2 significantly outperforms existing open-weight models like Qwen3-VL (35.5 vs 29.6 accuracy on video counting) and surpasses proprietary models like Gemini 3 Pro on some tasks (38.4 vs 20.0 F1 on video pointing and 56.2 vs 41.1 J&F on video tracking).

cs.CV↗

arXiv2026

MolmoPoint: Better Pointing for VLMs with Grounding Tokens

Grounding has become a fundamental capability of vision-language models (VLMs). Most existing VLMs point by generating coordinates as part of their text output, which requires learning a complicated coordinate system and results in a high token count. Instead, we propose a more intuitive pointing mechanism that directly selects the visual tokens that contain the target concept. Our model generates a special pointing token that cross-attends to the input image or video tokens and selects the appropriate one. To make this model more fine-grained, we follow these pointing tokens with an additional special token that selects a fine-grained subpatch within the initially selected region, and then a third token that specifies a location within that subpatch. We further show that performance improves by generating points sequentially in a consistent order, encoding the relative position of the previously selected point, and including a special no-more-points class when selecting visual tokens. Using this method, we set a new state-of-the-art on image pointing (70.7% on PointBench), set a new state-of-the-art among fully open models on GUI pointing (61.1% on ScreenSpotPro), and improve video pointing (59.1% human preference win rate vs. a text coordinate baseline) and tracking (+6.3% gain on Molmo2Track). We additionally show that our method achieves much higher sample efficiency and discuss the qualitative differences that emerge from this design change.

cs.CV↗

arXiv2026

RFDM: Residual Flow Diffusion Model for Efficient Causal Video Editing

Instructional video editing applies edits to an input video using only text prompts, enabling intuitive natural-language control. Despite rapid progress, most methods still require fixed-length inputs and substantial compute. Meanwhile, autoregressive video generation enables efficient variable-length synthesis, yet remains under-explored for video editing. We introduce a causal, efficient video editing model that edits variable-length videos frame by frame. For efficiency, we start from a 2D image-to-image (I2I) diffusion model and adapt it to video-to-video (V2V) editing by conditioning the edit at time step t on the model's prediction at t-1. To leverage videos' temporal redundancy, we propose a new I2I diffusion forward process formulation that encourages the model to predict the residual between the target output and the previous prediction. We call this Residual Flow Diffusion Model (RFDM), which focuses the denoising process on changes between consecutive frames. Moreover, we propose a new benchmark that better ranks state-of-the-art methods for editing tasks. Trained on paired video data for global/local style transfer and object removal, RFDM surpasses I2I-based methods and competes with fully spatiotemporal (3D) V2V models, while matching the compute of image models and scaling independently of input video length. More content can be found in: https://smsd75.github.io/RFDM_page/

cs.CV↗

arXiv2026

Self-supervised Learning of Echocardiographic Video Representations via Online Cluster Distillation

Self-supervised learning (SSL) has achieved major advances in natural images and video understanding, but challenges remain in domains like echocardiography (heart ultrasound) due to subtle anatomical structures, complex temporal dynamics, and the current lack of domain-specific pre-trained models. Existing SSL approaches such as contrastive, masked modeling, and clustering-based methods struggle with high intersample similarity, sensitivity to low PSNR inputs common in ultrasound, or aggressive augmentations that distort clinically relevant features. We present DISCOVR (Distilled Image Supervision for Cross Modal Video Representation), a self-supervised dual branch framework for cardiac ultrasound video representation learning. DISCOVR combines a clustering-based video encoder that models temporal dynamics with an online image encoder that extracts fine-grained spatial semantics. These branches are connected through a semantic cluster distillation loss that transfers anatomical knowledge from the evolving image encoder to the video encoder, enabling temporally coherent representations enriched with fine-grained semantic understanding.Evaluated on six echocardiography datasets spanning fetal, pediatric, and adult populations, DISCOVR outperforms both specialized video anomaly detection methods and state-of-the-art video-SSL baselines in zero-shot and linear probing setups,achieving superior segmentation transfer and strong downstream performance on clinically relevant tasks such as LVEF prediction. Code available at: https://github.com/mdivyanshu97/DISCOVR

cs.CV↗

arXiv2026

Evaluating Foundation Models' 3D Understanding Through Multi-View Correspondence Analysis

Benchmarking 3D spatial understanding of foundation models is essential for real-world applications such as robotics and autonomous driving. Existing evaluations often rely on downstream fine-tuning with linear heads or task-specific decoders, making it difficult to isolate the intrinsic 3D reasoning ability of pre-trained encoders. In this work, we introduce a novel benchmark for in-context 3D scene understanding that requires no fine-tuning and directly probes the quality of dense visual features. Building on the Hummingbird framework, which evaluates in-context 2D scene understanding, we extend the setup to the 3D Multi-View ImageNet (MVImgNet) dataset. Given a set of images depicting objects at specific camera angles (keys), we benchmark the performance of segmenting novel views (queries) and report the scores in 4 categories of easy, medium, hard, and extreme based on the key-query view contrast. We benchmark 7 state-of-the-art foundation models and show that DINO-based encoders remain competitive across large viewpoint shifts. Our code is publicly available at https://github.com/ToyeshC/open-hummingbird-3d-eval.

cs.CV↗

arXiv2025

MoSiC: Optimal-Transport Motion Trajectory for Dense Self-Supervised Learning

Dense self-supervised learning has shown great promise for learning pixel- and patch-level representations, but extending it to videos remains challenging due to the complexity of motion dynamics. Existing approaches struggle as they rely on static augmentations that fail under object deformations, occlusions, and camera movement, leading to inconsistent feature learning over time. We propose a motion-guided self-supervised learning framework that clusters dense point tracks to learn spatiotemporally consistent representations. By leveraging an off-the-shelf point tracker, we extract long-range motion trajectories and optimize feature clustering through a momentum-encoder-based optimal transport mechanism. To ensure temporal coherence, we propagate cluster assignments along tracked points, enforcing feature consistency across views despite viewpoint changes. Integrating motion as an implicit supervisory signal, our method learns representations that generalize across frames, improving robustness in dynamic scenes and challenging occlusion scenarios. By initializing from strong image-pretrained models and leveraging video data for training, we improve state-of-the-art by 1% to 6% on six image and video datasets and four evaluation benchmarks. The implementation is publicly available at our GitHub repository: https://github.com/SMSD75/MoSiC/tree/main

cs.CV↗

arXiv2025

REALEDIT: Reddit Edits As a Large-scale Empirical Dataset for Image Transformations

Existing image editing models struggle to meet real-world demands. Despite excelling in academic benchmarks, they have yet to be widely adopted for real user needs. Datasets that power these models use artificial edits, lacking the scale and ecological validity necessary to address the true diversity of user requests. We introduce REALEDIT, a large-scale image editing dataset with authentic user requests and human-made edits sourced from Reddit. REALEDIT includes a test set of 9300 examples to evaluate models on real user requests. Our results show that existing models fall short on these tasks, highlighting the need for realistic training data. To address this, we introduce 48K training examples and train our REALEDIT model, achieving substantial gains - outperforming competitors by up to 165 Elo points in human judgment and 92 percent relative improvement on the automated VIEScore metric. We deploy our model on Reddit, testing it on new requests, and receive positive feedback. Beyond image editing, we explore REALEDIT's potential in detecting edited images by partnering with a deepfake detection non-profit. Finetuning their model on REALEDIT data improves its F1-score by 14 percentage points, underscoring the dataset's value for broad applications.

cs.CV↗

arXiv2025

Near, far: Patch-ordering enhances vision foundation models' scene understanding

We introduce NeCo: Patch Neighbor Consistency, a novel self-supervised training loss that enforces patch-level nearest neighbor consistency across a student and teacher model. Compared to contrastive approaches that only yield binary learning signals, i.e., 'attract' and 'repel', this approach benefits from the more fine-grained learning signal of sorting spatially dense features relative to reference patches. Our method leverages differentiable sorting applied on top of pretrained representations, such as DINOv2-registers to bootstrap the learning signal and further improve upon them. This dense post-pretraining leads to superior performance across various models and datasets, despite requiring only 19 hours on a single GPU. This method generates high-quality dense feature encoders and establishes several new state-of-the-art results such as +5.5% and +6% for non-parametric in-context semantic segmentation on ADE20k and Pascal VOC, +7.2% and +5.7% for linear segmentation evaluations on COCO-Things and -Stuff and improvements in the 3D understanding of multi-view consistency on SPair-71k, by more than 1.5%.

cs.CV↗

arXiv2025

A Contrastive Teacher-Student Framework for Novelty Detection under Style Shifts

There have been several efforts to improve Novelty Detection (ND) performance. However, ND methods often suffer significant performance drops under minor distribution shifts caused by changes in the environment, known as style shifts. This challenge arises from the ND setup, where the absence of out-of-distribution (OOD) samples during training causes the detector to be biased toward the dominant style features in the in-distribution (ID) data. As a result, the model mistakenly learns to correlate style with core features, using this shortcut for detection. Robust ND is crucial for real-world applications like autonomous driving and medical imaging, where test samples may have different styles than the training data. Motivated by this, we propose a robust ND method that crafts an auxiliary OOD set with style features similar to the ID set but with different core features. Then, a task-based knowledge distillation strategy is utilized to distinguish core features from style features and help our model rely on core features for discriminating crafted OOD and ID sets. We verified the effectiveness of our method through extensive experimental evaluations on several datasets, including synthetic and real-world benchmarks, against nine different ND methods.

cs.CV↗

arXiv2024

Redefining Normal: A Novel Object-Level Approach for Multi-Object Novelty Detection

In the realm of novelty detection, accurately identifying outliers in data without specific class information poses a significant challenge. While current methods excel in single-object scenarios, they struggle with multi-object situations due to their focus on individual objects. Our paper suggests a novel approach: redefining `normal' at the object level in training datasets. Rather than the usual image-level view, we consider the most dominant object in a dataset as the norm, offering a perspective that is more effective for real-world scenarios. Adapting to our object-level definition of `normal', we modify knowledge distillation frameworks, where a student network learns from a pre-trained teacher network. Our first contribution, DeFeND(Dense Feature Fine-tuning on Normal Data), integrates dense feature fine-tuning into the distillation process, allowing the teacher network to focus on object-level features with a self-supervised loss. The second is masked knowledge distillation, where the student network works with partially hidden inputs, honing its ability to deduce and generalize from incomplete data. This approach not only fares well in single-object novelty detection but also considerably surpasses existing methods in multi-object contexts. The implementation is available at: https://github.com/SMSD75/Redefining_Normal_ACCV24/tree/main

cs.CV↗

arXiv2024

Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models

Today's most advanced vision-language models (VLMs) remain proprietary. The strongest open-weight models rely heavily on synthetic data from proprietary VLMs to achieve good performance, effectively distilling these closed VLMs into open ones. As a result, the community has been missing foundational knowledge about how to build performant VLMs from scratch. We present Molmo, a new family of VLMs that are state-of-the-art in their class of openness. Our key contribution is a collection of new datasets called PixMo, including a dataset of highly detailed image captions for pre-training, a free-form image Q&A dataset for fine-tuning, and an innovative 2D pointing dataset, all collected without the use of external VLMs. The success of our approach relies on careful modeling choices, a well-tuned training pipeline, and, most critically, the quality of our newly collected datasets. Our best-in-class 72B model not only outperforms others in the class of open weight and data models, but also outperforms larger proprietary models including Claude 3.5 Sonnet, and Gemini 1.5 Pro and Flash, second only to GPT-4o based on both academic benchmarks and on a large human evaluation. Our model weights, new datasets, and source code are available at https://molmo.allenai.org/blog.

cs.CV↗