Recent advances in generative diffusion models have shown a notable inherent understanding of image style and semantics. In this paper, we leverage the self-attention features from pretrained diffusion networks to transfer the visual characteristics from a reference to generated images. Unlike previous work that uses these features as plug-and-play attributes, we propose a novel attention distillation loss calculated between the ideal and current stylization results, based on which we optimize the synthesized image via backpropagation in latent space. Next, we propose an improved Classifier Guidance that integrates attention distillation loss into the denoising sampling process, further accelerating the synthesis and enabling a broad range of image generation applications. Extensive experiments have demonstrated the extraordinary performance of our approach in transferring the examples' style, appearance, and texture to new images in synthesis. Code is available at https://github.com/xugao97/AttentionDistillation.
We present StyleMotif, a novel Stylized Motion Latent Diffusion model, generating motion conditioned on both content and style from multiple modalities. Unlike existing approaches that either focus on generating diverse motion content or transferring style from sequences, StyleMotif seamlessly synthesizes motion across a wide range of content while incorporating stylistic cues from multi-modal inputs, including motion, text, image, video, and audio. To achieve this, we introduce a style-content cross fusion mechanism and align a style encoder with a pre-trained multi-modal model, ensuring that the generated motion accurately captures the reference style while preserving realism. Extensive experiments demonstrate that our framework surpasses existing methods in stylized motion generation and exhibits emergent capabilities for multi-modal motion stylization, enabling more nuanced motion synthesis. Source code and pre-trained models will be released upon acceptance. Project Page: https://stylemotif.github.io
Neural Metamorphosis (NeuMeta) is a recent paradigm for generating neural networks of varying width and depth. Based on Implicit Neural Representation (INR), NeuMeta learns a continuous weight manifold, enabling the direct generation of compressed models, including those with configurations not seen during training. While promising, the original formulation of NeuMeta proves effective only for the final layers of the undelying model, limiting its broader applicability. In this work, we propose a training algorithm that extends the capabilities of NeuMeta to enable full-network metamorphosis with minimal accuracy degradation. Our approach follows a structured recipe comprising block-wise incremental training, INR initialization, and strategies for replacing batch normalization. The resulting metamorphic networks maintain competitive accuracy across a wide range of compression ratios, offering a scalable solution for adaptable and efficient deployment of deep models. The code is available at: https://github.com/TSommariva/HTTY_NeuMeta.
Retrieval-Augmented Generation (RAG) addresses limitations of large language models (LLMs) by leveraging a vector database to provide more accurate and up-to-date information. When a user submits a query, RAG executes a vector search to find relevant documents, which are then used to generate a response. However, ensuring the relevance of retrieved documents with a query would be a big challenge. To address this, a secondary model, known as a relevant grader, can be served to verify its relevance. To reduce computational requirements of a relevant grader, a lightweight small language model is preferred. In this work, we finetuned llama-3.2-1b as a relevant grader and achieved a significant increase in precision from 0.1301 to 0.7750. Its precision is comparable to that of llama-3.1-70b. Our code is available at https://github.com/taeheej/Lightweight-Relevance-Grader-in-RAG.
Building models that comprehends videos and responds specific user instructions is a practical and challenging topic, as it requires mastery of both vision understanding and knowledge reasoning. Compared to language and image modalities, training efficiency remains a serious problem as existing studies train models on massive sparse videos paired with brief descriptions. In this paper, we introduce \textbf{VidCoM}, a fast adaptive framework that leverages Large Language Models (LLMs) to reason about videos using lightweight visual tools. Specifically, we reveal that the key to responding to specific instructions is focusing on relevant video events, and utilize two visual tools, structured scene graph generation and descriptive image caption generation, to gather and represent the event information. Thus, a LLM enriched with world knowledge is adopted as the reasoning agent to achieve the responses by performing multiple reasoning steps on specific video events. To address the difficulty of LLMs identifying video events, we further propose an Instruction-oriented Video Events Recognition (InsOVER) algorithm. This algorithm locates the corresponding video events based on an efficient Hungarian matching between decompositions of linguistic instructions and video events, thereby enabling LLMs to interact effectively with extended videos. Extensive experiments on two typical video comprehension tasks show that the proposed tuning-free framework outperforms the pre-trained models including Flamingo-80B, to achieve the state-of-the-art performance. Our source code and system will be publicly available.
While large language models (LLMs) have exhibited impressive instruction-following capabilities, it is still unclear whether and to what extent they can respond to explicit constraints that might be entailed in various instructions. As a significant aspect of LLM alignment, it is thus important to formulate such a specialized set of instructions as well as investigate the resulting behavior of LLMs. To address this vacancy, we propose a new benchmark CoDI-Eval to systematically and comprehensively evaluate LLMs' responses to instructions with various constraints. We construct a large collection of constraints-attributed instructions as a test suite focused on both generalization and coverage. Specifically, we advocate an instruction diversification process to synthesize diverse forms of constraint expression and also deliberate the candidate task taxonomy with even finer-grained sub-categories. Finally, we automate the entire evaluation process to facilitate further developments. Different from existing studies on controllable text generation, CoDI-Eval extends the scope to the prevalent instruction-following paradigm for the first time. We provide extensive evaluations of representative LLMs (e.g., ChatGPT, Vicuna) on CoDI-Eval, revealing their limitations in following instructions with specific constraints and there is still a significant gap between open-source and commercial closed-source LLMs. We believe this benchmark will facilitate research into improving the controllability of LLMs' responses to instructions. Our data and code are available at https://github.com/Xt-cyh/CoDI-Eval.
Human hands possess the dexterity to interact with diverse objects such as grasping specific parts of the objects and/or approaching them from desired directions. More importantly, humans can grasp objects of any shape without object-specific skills. Recent works synthesize grasping motions following single objectives such as a desired approach heading direction or a grasping area. Moreover, they usually rely on expensive 3D hand-object data during training and inference, which limits their capability to synthesize grasping motions for unseen objects at scale. In this paper, we unify the generation of hand-object grasping motions across multiple motion objectives, diverse object shapes and dexterous hand morphologies in a policy learning framework GraspXL. The objectives are composed of the graspable area, heading direction during approach, wrist rotation, and hand position. Without requiring any 3D hand-object interaction data, our policy trained with 58 objects can robustly synthesize diverse grasping motions for more than 500k unseen objects with a success rate of 82.2%. At the same time, the policy adheres to objectives, which enables the generation of diverse grasps per object. Moreover, we show that our framework can be deployed to different dexterous hands and work with reconstructed or generated objects. We quantitatively and qualitatively evaluate our method to show the efficacy of our approach. Our model, code, and the large-scale generated motions are available at https://eth-ait.github.io/graspxl/.
Drone detection has benefited from improvements in deep neural networks, but like many other applications, suffers from the availability of accurate data for training. Synthetic data provides a potential for low-cost data generation and has been shown to improve data availability and quality. However, models trained on synthetic datasets need to prove their ability to perform on real-world data, known as the problem of sim-to-real transferability. Here, we present a drone detection Faster-RCNN model trained on a purely synthetic dataset that transfers to real-world data. We found that it achieves an AP_50 of 97.0% when evaluated on the MAV-Vid - a real dataset of flying drones - compared with 97.8% for an equivalent model trained on real-world data. Our results show that using synthetic data for drone detection has the potential to reduce data collection costs and improve labelling quality. These findings could be a starting point for more elaborate synthetic drone datasets. For example, realistic recreations of specific scenarios could de-risk the dataset generation of safety-critical applications such as the detection of drones at airports. Further, synthetic data may enable reliable drone detection systems, which could benefit other areas, such as unmanned traffic management systems. The code is available https://github.com/mazqtpopx/cranfield-synthetic-drone-detection alongside the datasets https://huggingface.co/datasets/mazqtpopx/cranfield-synthetic-drone-detection.
Graph-based retrieval-augmented generation (GraphRAG) improves multi-hop question answering, but deployed systems face two costs that research comparisons rarely measure: building the graph, and removing documents from it. We present Lightfoot, a GraphRAG system in which every entity, link, fact and synonym edge is a deterministic function of per-passage extractions carrying chunk-level provenance. Indexing makes one language-model call per passage, retrieval makes none, and deleting a document is a database operation with no model calls. On the full 11,402-passage MuSiQue corpus we verify that, after deleting 10% of the documents, the graph is identical to a fresh index of the surviving documents. We compare Lightfoot with HippoRAG 2, LightRAG and fast-graphrag under identical extraction, answer and embedding models on MuSiQue, 2WikiMultiHopQA and HotpotQA. Lightfoot builds the cheapest index on all three datasets, 26-31% below HippoRAG 2 and 3-6x below LightRAG and fast-graphrag. Against HippoRAG 2 it is significantly better on MuSiQue retrieval (+2.8 points recall@5 and +4.3 recall@10 on a held-out split) and on HotpotQA answers (+7.7 points exact match), and significantly worse on 2WikiMultiHopQA (-3.9 recall@2, -8.4 F1). Deleting the same documents took LightRAG 2,187 model calls and 3.7 hours; fast-graphrag has no deletion path. Code and per-question results: https://github.com/udayallu-ml/Lightfoot-GraphRAG
The rapid rise of video diffusion models has enabled the generation of highly realistic and temporally coherent videos, raising critical concerns about content authenticity, provenance, and misuse. Existing watermarking approaches, whether passive, post-hoc, or adapted from image-based techniques, often struggle to withstand video-specific manipulations such as frame insertion, dropping, or reordering, and typically degrade visual quality. In this work, we introduce VIDSTAMP, a watermarking framework that embeds per-frame or per-segment messages directly into the latent space of temporally-aware video diffusion models. By fine-tuning the model's decoder through a two-stage pipeline, first on static image datasets to promote spatial message separation, and then on synthesized video sequences to restore temporal consistency, VIDSTAMP learns to embed high-capacity, flexible watermarks with minimal perceptual impact. Leveraging architectural components such as 3D convolutions and temporal attention, our method imposes no additional inference cost and offers better perceptual quality than prior methods, while maintaining comparable robustness against common distortions and tampering. VIDSTAMP embeds 768 bits per video (48 bits per frame) with a bit accuracy of 95.0%, achieves a log P-value of -166.65 (lower is better), and maintains a video quality score of 0.836, comparable to unwatermarked outputs (0.838) and surpassing prior methods in capacity-quality tradeoffs. Code: Code: \url{https://github.com/SPIN-UMass/VidStamp}
Diffusion models have recently motivated great success in many generation tasks like object removal. Nevertheless, existing image decomposition methods struggle to disentangle semi-transparent or transparent layer occlusions due to mask prior dependencies, static object assumptions, and the lack of datasets. In this paper, we delve into a novel task: Layer-Wise Decomposition of Alpha-Composited Images, aiming to recover constituent layers from single overlapped images under the condition of semi-transparent/transparent alpha layer non-linear occlusion. To address challenges in layer ambiguity, generalization, and data scarcity, we first introduce AlphaBlend, the first large-scale and high-quality dataset for transparent and semi-transparent layer decomposition, supporting six real-world subtasks (e.g., translucent flare removal, semi-transparent cell decomposition, glassware decomposition). Building on this dataset, we present DiffDecompose, a diffusion Transformer-based framework that learns the posterior over possible layer decompositions conditioned on the input image, semantic prompts, and blending type. Rather than regressing alpha mattes directly, DiffDecompose performs In-Context Decomposition, enabling the model to predict one or multiple layers without per-layer supervision, and introduces Layer Position Encoding Cloning to maintain pixel-level correspondence across layers. Extensive experiments on the proposed AlphaBlend dataset and public LOGO dataset verify the effectiveness of DiffDecompose. The code and dataset will be available upon paper acceptance. Our code will be available at: https://github.com/Wangzt1121/DiffDecompose.
The AI for Nuclear Energy workshop at Oak Ridge National Laboratory evaluated the potential of Large Language Models (LLMs) to accelerate fusion and fission research. Fourteen interdisciplinary teams explored diverse nuclear science challenges using ChatGPT, Gemini, Claude, and other AI models over a single day. Applications ranged from developing foundation models for fusion reactor control to automating Monte Carlo simulations, predicting material degradation, and designing experimental programs for advanced reactors. Teams employed structured workflows combining prompt engineering, deep research capabilities, and iterative refinement to generate hypotheses, prototype code, and research strategies. Key findings demonstrate that LLMs excel at early-stage exploration, literature synthesis, and workflow design, successfully identifying research gaps and generating plausible experimental frameworks. However, significant limitations emerged, including difficulties with novel materials designs, advanced code generation for modeling and simulation, and domain-specific details requiring expert validation. The successful outcomes resulted from expert-driven prompt engineering and treating AI as a complementary tool rather than a replacement for physics-based methods. The workshop validated AI's potential to accelerate nuclear energy research through rapid iteration and cross-disciplinary synthesis while highlighting the need for curated nuclear-specific datasets, workflow automation, and specialized model development. These results provide a roadmap for integrating AI tools into nuclear science workflows, potentially reducing development cycles for safer, more efficient nuclear energy systems while maintaining rigorous scientific standards.
Chain-of-Thought (CoT) guides large language models (LLMs) to reason step-by-step, and can motivate their logical reasoning ability. While effective for logical tasks, CoT is not conducive to creative problem-solving which often requires out-of-box thoughts and is crucial for innovation advancements. In this paper, we explore the Leap-of-Thought (LoT) abilities within LLMs -- a non-sequential, creative paradigm involving strong associations and knowledge leaps. To this end, we study LLMs on the popular Oogiri game which needs participants to have good creativity and strong associative thinking for responding unexpectedly and humorously to the given image, text, or both, and thus is suitable for LoT study. Then to investigate LLMs' LoT ability in the Oogiri game, we first build a multimodal and multilingual Oogiri-GO dataset which contains over 130,000 samples from the Oogiri game, and observe the insufficient LoT ability or failures of most existing LLMs on the Oogiri game. Accordingly, we introduce a creative Leap-of-Thought (CLoT) paradigm to improve LLM's LoT ability. CLoT first formulates the Oogiri-GO dataset into LoT-oriented instruction tuning data to train pretrained LLM for achieving certain LoT humor generation and discrimination abilities. Then CLoT designs an explorative self-refinement that encourages the LLM to generate more creative LoT data via exploring parallels between seemingly unrelated concepts and selects high-quality data to train itself for self-refinement. CLoT not only excels in humor generation in the Oogiri game but also boosts creative abilities in various tasks like cloud guessing game and divergent association task. These findings advance our understanding and offer a pathway to improve LLMs' creative capacities for innovative applications across domains. The dataset, code, and models will be released online. https://zhongshsh.github.io/CLoT/.
The fourth-generation Wireless Technology (4G) has been adopted by all major operators in the world and has already ruled the cellular landscape for around a decade. A lot of researches and new technologies are being considered as potential elements contributing to the next generation wireless communication (5G). The lack of realistic and flexible experimentation platforms for collecting real communication data has limited and slowed the landing of new approaches. Software Defined Radio (SDR) can provide flexible, upgradable, and long lifetime radio equipment for the wireless communications infrastructure, which can also provide more flexible and possibly cheaper multi-standard-terminals for end users. By altering the open-source code individually, we can freely perform the real value measurement. This paper provides a real Long Term Evolution (LTE) channel measurement method based on the OpenAirInterface (OAI) for the evaluation of the channel prediction algorithm. Firstly, the experimentation platform will be established by using OAI, Universal Software Radio Peripheral (USRP), and commercial User Equipment (UE). Then, some source codes of OAI are analyzed and changed, so that the real-time over-the-air channel measurement can be achieved. The results from the measurement are then trained and tested on the channel prediction algorithm. The results of the test illustrate that the implemented channel measurement method can meet the need for algorithms' verification and can be further extended for more development of algorithms.
In image denoising networks, feature scaling is widely used to enlarge the receptive field size and reduce computational costs. This practice, however, also leads to the loss of high-frequency information and fails to consider within-scale characteristics. Recently, dynamic convolution has exhibited powerful capabilities in processing high-frequency information (e.g., edges, corners, textures), but previous works lack sufficient spatial contextual information in filter generation. To alleviate these issues, we propose to employ dynamic convolution to improve the learning of high-frequency and multi-scale features. Specifically, we design a spatially enhanced kernel generation (SEKG) module to improve dynamic convolution, enabling the learning of spatial context information with a very low computational complexity. Based on the SEKG module, we propose a dynamic convolution block (DCB) and a multi-scale dynamic convolution block (MDCB). The former enhances the high-frequency information via dynamic convolution and preserves low-frequency information via skip connections. The latter utilizes shared adaptive dynamic kernels and the idea of dilated convolution to achieve efficient multi-scale feature extraction. The proposed multi-dimension feature integration (MFI) mechanism further fuses the multi-scale features, providing precise and contextually enriched feature representations. Finally, we build an efficient denoising network with the proposed DCB and MDCB, named ADFNet. It achieves better performance with low computational complexity on real-world and synthetic Gaussian noisy datasets. The source code is available at https://github.com/it-hao/ADFNet.
Medical report generation (MRG) is essential for computer-aided diagnosis and medication guidance, which can relieve the heavy burden of radiologists by automatically generating the corresponding medical reports according to the given radiology image. However, due to the spurious correlations within image-text data induced by visual and linguistic biases, it is challenging to generate accurate reports reliably describing lesion areas. Moreover, the cross-modal confounders are usually unobservable and challenging to be eliminated explicitly. In this paper, we aim to mitigate the cross-modal data bias for MRG from a new perspective, i.e., cross-modal causal intervention, and propose a novel Visual-Linguistic Causal Intervention (VLCI) framework for MRG, which consists of a visual deconfounding module (VDM) and a linguistic deconfounding module (LDM), to implicitly mitigate the visual-linguistic confounders by causal front-door intervention. Specifically, due to the absence of a generalized semantic extractor, the VDM explores and disentangles the visual confounders from the patch-based local and global features without expensive fine-grained annotations. Simultaneously, due to the lack of knowledge encompassing the entire field of medicine, the LDM eliminates the linguistic confounders caused by salient visual features and high-frequency context without constructing a terminology database. Extensive experiments on IU-Xray and MIMIC-CXR datasets show that our VLCI significantly outperforms the state-of-the-art MRG methods. The code and models are available at https://github.com/WissingChen/VLCI.
The development of large language models (LLMs) has significantly enhanced the capabilities of multimodal LLMs (MLLMs) as general assistants. However, lack of user-specific knowledge still restricts their application in human's daily life. In this paper, we introduce the Retrieval Augmented Personalization (RAP) framework for MLLMs' personalization. Starting from a general MLLM, we turn it into a personalized assistant in three steps. (a) Remember: We design a key-value database to store user-related information, e.g., user's name, avatar and other attributes. (b) Retrieve: When the user initiates a conversation, RAP will retrieve relevant information from the database using a multimodal retriever. (c) Generate: The input query and retrieved concepts' information are fed into MLLMs to generate personalized, knowledge-augmented responses. Unlike previous methods, RAP allows real-time concept editing via updating the external database. To further improve generation quality and alignment with user-specific information, we design a pipeline for data collection and create a specialized dataset for personalized training of MLLMs. Based on the dataset, we train a series of MLLMs as personalized multimodal assistants. By pretraining on large-scale dataset, RAP-MLLMs can generalize to infinite visual concepts without additional finetuning. Our models demonstrate outstanding flexibility and generation quality across a variety of tasks, such as personalized image captioning, question answering and visual recognition. The code, data and models are available at https://hoar012.github.io/RAP-Project/.
Artificial Intelligence (AI) has demonstrated significant potential in healthcare, particularly in disease diagnosis and treatment planning. Recent progress in Medical Large Vision-Language Models (Med-LVLMs) has opened up new possibilities for interactive diagnostic tools. However, these models often suffer from factual hallucination, which can lead to incorrect diagnoses. Fine-tuning and retrieval-augmented generation (RAG) have emerged as methods to address these issues. However, the amount of high-quality data and distribution shifts between training data and deployment data limit the application of fine-tuning methods. Although RAG is lightweight and effective, existing RAG-based approaches are not sufficiently general to different medical domains and can potentially cause misalignment issues, both between modalities and between the model and the ground truth. In this paper, we propose a versatile multimodal RAG system, MMed-RAG, designed to enhance the factuality of Med-LVLMs. Our approach introduces a domain-aware retrieval mechanism, an adaptive retrieved contexts selection method, and a provable RAG-based preference fine-tuning strategy. These innovations make the RAG process sufficiently general and reliable, significantly improving alignment when introducing retrieved contexts. Experimental results across five medical datasets (involving radiology, ophthalmology, pathology) on medical VQA and report generation demonstrate that MMed-RAG can achieve an average improvement of 43.8% in the factual accuracy of Med-LVLMs. Our data and code are available in https://github.com/richard-peng-xia/MMed-RAG.
Retrieval-augmented Generation (RAG) enhances Large Language Models (LLMs) by integrating external knowledge to reduce hallucinations and incorporate up-to-date information without retraining. As an essential part of RAG, external knowledge bases are commonly built by extracting structured data from unstructured PDF documents using Optical Character Recognition (OCR). However, given the imperfect prediction of OCR and the inherent non-uniform representation of structured data, knowledge bases inevitably contain various OCR noises. In this paper, we introduce OHRBench, the first benchmark for understanding the cascading impact of OCR on RAG systems. OHRBench includes 350 carefully selected unstructured PDF documents from six real-world RAG application domains, along with Q&As derived from multimodal elements in documents, challenging existing OCR solutions used for RAG To better understand OCR's impact on RAG systems, we identify two primary types of OCR noise: Semantic Noise and Formatting Noise and apply perturbation to generate a set of structured data with varying degrees of each OCR noise. Using OHRBench, we first conduct a comprehensive evaluation of current OCR solutions and reveal that none is competent for constructing high-quality knowledge bases for RAG systems. We then systematically evaluate the impact of these two noise types and demonstrate the vulnerability of RAG systems. Furthermore, we discuss the potential of employing Vision-Language Models (VLMs) without OCR in RAG systems. Code: https://github.com/opendatalab/OHR-Bench
Foundational models with billions of parameters which have been trained on large corpora of data have demonstrated non-trivial skills in a variety of domains. However, due to their monolithic structure, it is challenging and expensive to augment them or impart new skills. On the other hand, due to their adaptation abilities, several new instances of these models are being trained towards new domains and tasks. In this work, we study the problem of efficient and practical composition of existing foundation models with more specific models to enable newer capabilities. To this end, we propose CALM -- Composition to Augment Language Models -- which introduces cross-attention between models to compose their representations and enable new capabilities. Salient features of CALM are: (i) Scales up LLMs on new tasks by 're-using' existing LLMs along with a few additional parameters and data, (ii) Existing model weights are kept intact, and hence preserves existing capabilities, and (iii) Applies to diverse domains and settings. We illustrate that augmenting PaLM2-S with a smaller model trained on low-resource languages results in an absolute improvement of up to 13\% on tasks like translation into English and arithmetic reasoning for low-resource languages. Similarly, when PaLM2-S is augmented with a code-specific model, we see a relative improvement of 40\% over the base model for code generation and explanation tasks -- on-par with fully fine-tuned counterparts.