Animating hand-drawn sketches using traditional tools is challenging and complex. Sketches provide a visual basis for explanations, and animating these sketches offers an experience of real-time scenarios. We propose an approach for animating a given input sketch based on a descriptive text prompt. Our method utilizes a parametric representation of the sketch's strokes. Unlike previous methods, which struggle to estimate smooth and accurate motion and often fail to preserve the sketch's topology, we leverage a pre-trained text-to-video diffusion model with SDS loss to guide the motion of the sketch's strokes. We introduce length-area (LA) regularization to ensure temporal consistency by accurately estimating the smooth displacement of control points across the frame sequence. Additionally, to preserve shape and avoid topology changes, we apply a shape-preserving As-Rigid-As-Possible (ARAP) loss to maintain sketch rigidity. Our method surpasses state-of-the-art performance in both quantitative and qualitative evaluations.
To guide a learner in mastering action skills, it is crucial for a coach to 1) reason through the learner's action execution and technical points (TechPoints), and 2) provide detailed, comprehensible feedback on what is done well and what can be improved. However, existing score-based action assessment methods are still far from reaching this practical scenario. To bridge this gap, we investigate a new task termed Descriptive Action Coaching (DescCoach) which requires the model to provide detailed commentary on what is done well and what can be improved beyond a simple quality score for action execution. To this end, we first build a new dataset named EE4D-DescCoach. Through an automatic annotation pipeline, our dataset goes beyond the existing action assessment datasets by providing detailed TechPoint-level commentary. Furthermore, we propose TechCoach, a new framework that explicitly incorporates TechPoint-level reasoning into the DescCoach process. The central to our method lies in the Context-aware TechPoint Reasoner, which enables TechCoach to learn TechPoint-related quality representation by querying visual context under the supervision of TechPoint-level coaching commentary. By leveraging the visual context and the TechPoint-related quality representation, a unified TechPoint-aware Action Assessor is then employed to provide the overall coaching commentary together with the quality score. Combining all of these, we establish a new benchmark for DescCoach and evaluate the effectiveness of our method through extensive experiments. The data and code will be made publicly available.
Vision-Language Models (VLMs) have recently demonstrated remarkable capabilities in comprehending complex visual content. However, the mechanisms underlying how VLMs process visual information remain largely unexplored. In this paper, we conduct a thorough empirical analysis, focusing on attention modules across layers. We reveal several key insights about how these models process visual data: (i) the internal representation of the query tokens (e.g., representations of "describe the image"), is utilized by VLMs to store global image information; we demonstrate that these models generate surprisingly descriptive responses solely from these tokens, without direct access to image tokens. (ii) Cross-modal information flow is predominantly influenced by the middle layers (approximately 25% of all layers), while early and late layers contribute only marginally.(iii) Fine-grained visual attributes and object details are directly extracted from image tokens in a spatially localized manner, i.e., the generated tokens associated with a specific object or attribute attend strongly to their corresponding regions in the image. We propose novel quantitative evaluation to validate our observations, leveraging real-world complex visual scenes. Finally, we demonstrate the potential of our findings in facilitating efficient visual processing in state-of-the-art VLMs.
Recent advances in language modeling have demonstrated significant improvements in zero-shot capabilities, including in-context learning, instruction following, and machine translation for extremely under-resourced languages (Tanzer et al., 2024). However, many languages with limited written resources rely primarily on formal descriptions of grammar and vocabulary. In this paper, we introduce a set of benchmarks to evaluate how well models can extract and classify information from the complex descriptions found in linguistic grammars. We present a Retrieval-Augmented Generation (RAG)-based approach that leverages these descriptions for downstream tasks such as machine translation. Our benchmarks encompass linguistic descriptions for 248 languages across 142 language families, focusing on typological features from WALS and Grambank. This set of benchmarks offers the first comprehensive evaluation of language models' in-context ability to accurately interpret and extract linguistic features, providing a critical resource for scaling NLP to low-resource languages. The code and data are publicly available at \url{https://github.com/al-the-eigenvalue/RAG-on-grammars}.
Maize farming is a major livelihood activity for many farmers in Ghana. Unfortunately, farmers usually do not obtain the expected returns on their investment due to reliance on rudimentary, labor-intensive, and inefficient methods of production. Using cross-sectional data from 359 maize farmers, this study investigates the profitability and determinants of the use of tractor services for maize production in Ejura-Sekyedumase, Ashanti Region of Ghana. Results from descriptive and profitability analyses reveal that tractor services such as ploughing and shelling are widely used, but their profitability varies significantly among farmers. Key factors influencing profitability include farm size, fertilizer quantity applied, and farmer experience. Results from a multivariate probit analysis also showed that farming experience, fertilizer quantity, and profit per acre have a positive influence on tractor service use for shelling, while household size, farm size, and FBO have a negative influence. Farming experience, fertilizer quantity, and profit per acre positively influence tractor service use for ploughing, while farm size has a negative influence. A t-test result reveals a statistically significant difference in profit between farmers who use tractor services and those who do not. Specifically, farmers who utilize tractor services on their maize farm had a return to cost of 9 percent more than those who do not (p-value < 0.05). The Kendall's result showed a moderate agreement among the maize farmers (first ranked being financial issues) in their ability to access/utilize tractor services on their farm.
We study the distributional properties of linear neural networks with random parameters in the context of large networks, where the number of layers diverges in proportion to the number of neurons per layer. Prior works have shown that in the infinite-width regime, where the number of neurons per layer grows to infinity while the depth remains fixed, neural networks converge to a Gaussian process, known as the Neural Network Gaussian Process. However, this Gaussian limit sacrifices descriptive power, as it lacks the ability to learn dependent features and produce output correlations that reflect observed labels. Motivated by these limitations, we explore the joint proportional limit in which both depth and width diverge but maintain a constant ratio, yielding a non-Gaussian distribution that retains correlations between outputs. Our contribution extends previous works by rigorously characterizing, for linear activation functions, the limiting distribution as a nontrivial mixture of Gaussians.
We present the Explabox: an open-source toolkit for transparent and responsible machine learning (ML) model development and usage. Explabox aids in achieving explainable, fair and robust models by employing a four-step strategy: explore, examine, explain and expose. These steps offer model-agnostic analyses that transform complex 'ingestibles' (models and data) into interpretable 'digestibles'. The toolkit encompasses digestibles for descriptive statistics, performance metrics, model behavior explanations (local and global), and robustness, security, and fairness assessments. Implemented in Python, Explabox supports multiple interaction modes and builds on open-source packages. It empowers model developers and testers to operationalize explainability, fairness, auditability, and security. The initial release focuses on text data and models, with plans for expansion. Explabox's code and documentation are available open-source at https://explabox.readthedocs.io/.
Despite advances in Large Multi-modal Models, applying them to long and untrimmed video content remains challenging due to limitations in context length and substantial memory overhead. These constraints often lead to significant information loss and reduced relevance in the model responses. With the exponential growth of video data across web platforms, understanding long-form video is crucial for advancing generalized intelligence. In this paper, we introduce SALOVA: Segment-Augmented LOng Video Assistant, a novel video-LLM framework designed to enhance the comprehension of lengthy video content through targeted retrieval process. We address two main challenges to achieve it: (i) We present the SceneWalk dataset, a high-quality collection of 87.8K long videos, each densely captioned at the segment level to enable models to capture scene continuity and maintain rich descriptive context. (ii) We develop robust architectural designs integrating dynamic routing mechanism and spatio-temporal projector to efficiently retrieve and process relevant video segments based on user queries. Our framework mitigates the limitations of current video-LMMs by allowing for precise identification and retrieval of relevant video segments in response to queries, thereby improving the contextual relevance of the generated responses. Through extensive experiments, SALOVA demonstrates enhanced capability in processing complex long-form videos, showing significant capability to maintain contextual integrity across extended sequences.
Molecule discovery is a pivotal research field, impacting everything from the medicines we take to the materials we use. Recently, Large Language Models (LLMs) have been widely adopted in molecule understanding and generation, yet the alignments between molecules and their corresponding captions remain a significant challenge. Previous endeavours often treat the molecule as a general SMILES string or molecular graph, neglecting the fine-grained alignments between the molecular sub-structures and the descriptive textual phrases, which are crucial for accurate and explainable predictions. In this case, we introduce MolReFlect, a novel teacher-student framework designed to contextually perform the molecule-caption alignments in a fine-grained way. Our approach initially leverages a larger teacher LLM to label the detailed alignments by directly extracting critical phrases from molecule captions or SMILES strings and implying them to corresponding sub-structures or characteristics. To refine these alignments, we propose In-Context Selective Reflection, which retrieves previous extraction results as context examples for teacher LLM to reflect and lets a smaller student LLM select from in-context reflection and previous extraction results. Finally, we enhance the learning process of the student LLM through Chain-of-Thought In-Context Molecule Tuning, integrating the fine-grained alignments and the reasoning processes within the Chain-of-Thought format. Our experimental results demonstrate that MolReFlect enables LLMs like Mistral-7B to significantly outperform the previous baselines, achieving SOTA performance on the ChEBI-20 dataset. This advancement not only enhances the generative capabilities of LLMs in the molecule-caption translation task, but also contributes to a more explainable framework.
Existing Image Manipulation Localization (IML) methods mostly rely heavily on task-specific designs, making them perform well only on the target IML task, while joint training on multiple IML tasks causes significant performance degradation, hindering real applications. To this end, we propose Omni-IML, the first generalist model designed to unify IML across diverse tasks. Specifically, Omni-IML achieves generalization through three key components: (1) a Modal Gate Encoder, which adaptively selects the optimal encoding modality per sample, (2) a Dynamic Weight Decoder, which dynamically adjusts decoder filters to the task at hand, and (3) an Anomaly Enhancement module that leverages box supervision to highlight the tampered regions and facilitate the learning of task-agnostic features. Beyond localization, to support interpretation of the tampered images, we construct Omni-273k, a large high-quality dataset that includes natural language descriptions of tampered artifact. It is annotated through our automatic, chain-of-thoughts annotation technique. We also design a simple-yet-effective interpretation module to better utilize these descriptive annotations. Our extensive experiments show that our single Omni-IML model achieves state-of-the-art performance across all four major IML tasks, providing a valuable solution for practical deployment and a promising direction of generalist models in image forensics. Our code and dataset will be publicly available.
Medical imaging has significantly revolutionized medical diagnostics and treatment planning, progressing from early X-ray usage to sophisticated methods like MRIs, CT scans, and ultrasounds. This paper investigates the use of deep learning for medical image captioning, with a particular focus on uterine ultrasound images. These images are vital in obstetrics and gynecology for diagnosing and monitoring various conditions across different age groups. However, their interpretation is often challenging due to their complexity and variability. To address this, a deep learning-based medical image captioning system was developed, integrating Convolutional Neural Networks with a Bidirectional Gated Recurrent Unit network. This hybrid model processes both image and text features to generate descriptive captions for uterine ultrasound images. Our experimental results demonstrate the effectiveness of this approach over baseline methods, with the proposed model achieving superior performance in generating accurate and informative captions, as indicated by higher BLEU and ROUGE scores. By enhancing the interpretation of uterine ultrasound images, our research aims to assist medical professionals in making timely and accurate diagnoses, ultimately contributing to improved patient care.
Multimodal sensing systems are increasingly prevalent in various real-world applications. Most existing multimodal learning approaches heavily rely on training with a large amount of complete multimodal data. However, such a setting is impractical in real-world IoT sensing applications where data is typically collected by distributed nodes with heterogeneous data modalities, and is also rarely labeled. In this paper, we propose MMBind, a new framework for multimodal learning on distributed and heterogeneous IoT data. The key idea of MMBind is to construct a pseudo-paired multimodal dataset for model training by binding data from disparate sources and incomplete modalities through a sufficiently descriptive shared modality. We demonstrate that data of different modalities observing similar events, even captured at different times and locations, can be effectively used for multimodal training. Moreover, we propose an adaptive multimodal learning architecture capable of training models with heterogeneous modality combinations, coupled with a weighted contrastive learning approach to handle domain shifts among disparate data. Evaluations on ten real-world multimodal datasets highlight that MMBind outperforms state-of-the-art baselines under varying data incompleteness and domain shift, and holds promise for advancing multimodal foundation model training in IoT applications.
Accurate segmentation of Optical Coherence Tomography (OCT) images is crucial for diagnosing and monitoring retinal diseases. However, the labor-intensive nature of pixel-level annotation limits the scalability of supervised learning with large datasets. Weakly Supervised Semantic Segmentation (WSSS) provides a promising alternative by leveraging image-level labels. In this study, we propose a novel WSSS approach that integrates structural guidance with text-driven strategies to generate high-quality pseudo labels, significantly improving segmentation performance. In terms of visual information, our method employs two processing modules that exchange raw image features and structural features from OCT images, guiding the model to identify where lesions are likely to occur. In terms of textual information, we utilize large-scale pretrained models from cross-domain sources to implement label-informed textual guidance and synthetic descriptive integration with two textual processing modules that combine local semantic features with consistent synthetic descriptions. By fusing these visual and textual components within a multimodal framework, our approach enhances lesion localization accuracy. Experimental results on three OCT datasets demonstrate that our method achieves state-of-the-art performance, highlighting its potential to improve diagnostic accuracy and efficiency in medical imaging.
The rapid spread of digital technologies has produced data-driven feedback loops, wearable devices, social media networks, and mobile applications that shape user behavior, motivation, and mental well-being. While these systems encourage self-improvement and the development of healthier habits through real-time feedback, they also create psychological risks such as technostress, addiction, and loss of autonomy. The present study also aims to investigate the positive and negative psychological consequences of feedback mechanisms on users' behaviour and well-being. Employing a descriptive survey method, the study collected data from 200 purposely selected users to assess changes in behaviour, motivation, and mental well-being related to health, social, and lifestyle applications. Results indicate that while feedback mechanisms facilitate goal attainment and social interconnection through streaks and badges, among other components, they also enhance anxiety, mental weariness, and loss of productivity due to actions that are considered feedback-seeking. Furthermore, test subjects reported that their actions are unconsciously shaped by app feedback, often at the expense of personal autonomy, while real-time feedback minimally influences professional or social interactions. The study shows that data-driven feedback loops deliver not only motivational benefits but also psychological challenges. To mitigate these risks, users should establish boundaries regarding their use of technology to prevent burnout and addiction, while developers need to refine feedback mechanisms to reduce cognitive load and foster more inclusive participation. Future research should focus on designing feedback mechanisms that promote well-being without compromising individual freedom or increasing social comparison.
Multimodal Large Language Models (MLLMs) excel at descriptive tasks within images but often struggle with precise object localization, a critical element for reliable visual interpretation. In contrast, traditional object detection models provide high localization accuracy but frequently generate detections lacking contextual coherence due to limited modeling of inter-object relationships. To address this fundamental limitation, we introduce the \textbf{Visual-Linguistic Agent (VLA), a collaborative framework that combines the relational reasoning strengths of MLLMs with the precise localization capabilities of traditional object detectors. In the VLA paradigm, the MLLM serves as a central Linguistic Agent, working collaboratively with specialized Vision Agents for object detection and classification. The Linguistic Agent evaluates and refines detections by reasoning over spatial and contextual relationships among objects, while the classification Vision Agent offers corrective feedback to improve classification accuracy. This collaborative approach enables VLA to significantly enhance both spatial reasoning and object localization, addressing key challenges in multimodal understanding. Extensive evaluations on the COCO dataset demonstrate substantial performance improvements across multiple detection models, highlighting VLA's potential to set a new benchmark in accurate and contextually coherent object detection.
In this paper, we introduce Motion-Grounded Video Reasoning, a new motion understanding task that requires generating visual answers (video segmentation masks) according to the input question, and hence needs implicit spatiotemporal reasoning and grounding. This task extends existing spatiotemporal grounding work focusing on explicit action/motion grounding, to a more general format by enabling implicit reasoning via questions. To facilitate the development of the new task, we collect a large-scale dataset called GROUNDMORE, which comprises 1,715 video clips, 249K object masks that are deliberately designed with 4 question types (Causal, Sequential, Counterfactual, and Descriptive) for benchmarking deep and comprehensive motion reasoning abilities. GROUNDMORE uniquely requires models to generate visual answers, providing a more concrete and visually interpretable response than plain texts. It evaluates models on both spatiotemporal grounding and reasoning, fostering to address complex challenges in motion-related video reasoning, temporal perception, and pixel-level understanding. Furthermore, we introduce a novel baseline model named Motion-Grounded Video Reasoning Assistant (MORA). MORA incorporates the multimodal reasoning ability from the Multimodal LLM, the pixel-level perception capability from the grounding model (SAM), and the temporal perception ability from a lightweight localization head. MORA achieves respectable performance on GROUNDMORE outperforming the best existing visual grounding baseline model by an average of 21.5% relatively. We hope this novel and challenging task will pave the way for future advancements in robust and general motion understanding via video reasoning segmentation
Recent research increasingly focuses on training vision-language models (VLMs) with long, detailed image captions. However, small-scale VLMs often struggle to balance the richness of these captions with the risk of hallucinating content during fine-tuning. In this paper, we explore how well VLMs adapt to such captions. To quantify caption quality, we propose Decomposed NLI (DNLI), an evaluation framework that breaks down generated captions into individual propositions, assessing each in isolation. This fine-grained analysis reveals a critical balance between capturing descriptive details and preventing hallucinations. Our findings show that simply reducing caption complexity or employing standard data curation techniques does not effectively resolve this issue. To tackle this challenge, we introduce Knowledge Adapted (KnowAda) fine-tuning, a data-centric approach that automatically adapts training data with the model's existing knowledge and visual understanding. KnowAda minimizes hallucinations while preserving high descriptiveness. We validate this approach across several small-scale VLMs (up to 7B parameters) and dense caption datasets, demonstrating that KnowAda effectively balances hallucination reduction and descriptiveness. Our results show that KnowAda outperforms various baselines in both automatic metrics and human evaluations. We will release our code and models.
We address the problem of asset pricing in a market where there is no risky asset. Previous work developed a theoretical model for a shadow riskless rate (SRR) for such a market in terms of the drift component of the state-price deflator for that asset universe. Assuming asset prices are modeled by correlated geometric Brownian motion, in this work we develop a computational approach to estimate the SRR from empirical datasets. The approach employs: principal component analysis to model the effects of the individual Brownian motions; singular value decomposition to capture the abrupt changes in condition number of the linear system whose solution provides the SRR values; and a regularization to control the rate of change of the condition number. Among other uses (e.g., for option pricing, developing a term structure of interest rate), the SRR can be employed as an investment discriminator between asset classes. We apply the computational procedure to markets consisting of groups of stocks, varying asset type and number. The theoretical and computational analysis provides not only the drift, but also the total volatility of the state-price deflator. We investigate the time trajectory of these two descriptive components of the state-price deflator for the empirical datasets.
We focus on collaborative and federated black-box optimization (BBOpt), where agents optimize their heterogeneous black-box functions through collaborative sequential experimentation. From a Bayesian optimization perspective, we address the fundamental challenges of distributed experimentation, heterogeneity, and privacy within BBOpt, and propose three unifying frameworks to tackle these issues: (i) a global framework where experiments are centrally coordinated, (ii) a local framework that allows agents to make decisions based on minimal shared information, and (iii) a predictive framework that enhances local surrogates through collaboration to improve decision-making. We categorize existing methods within these frameworks and highlight key open questions to unlock the full potential of federated BBOpt. Our overarching goal is to shift federated learning from its predominantly descriptive/predictive paradigm to a prescriptive one, particularly in the context of BBOpt - an inherently sequential decision-making problem.
We introduce BLIP3-KALE, a dataset of 218 million image-text pairs that bridges the gap between descriptive synthetic captions and factual web-scale alt-text. KALE augments synthetic dense image captions with web-scale alt-text to generate factually grounded image captions. Our two-stage approach leverages large vision-language models and language models to create knowledge-augmented captions, which are then used to train a specialized VLM for scaling up the dataset. We train vision-language models on KALE and demonstrate improvements on vision-language tasks. Our experiments show the utility of KALE for training more capable and knowledgeable multimodal models. We release the KALE dataset at https://huggingface.co/datasets/Salesforce/blip3-kale