Search

There has been immense progress recently in the visual quality of Stable Diffusion-based Super Resolution (SD-SR). However, deploying large diffusion models on computationally restricted devices such as mobile phones remains impractical due to the large model size and high latency. This is compounded for SR as it often operates at high res (e.g. 4Kx3K). In this work, we introduce Edge-SD-SR, the first parameter efficient and low latency diffusion model for image super-resolution. Edge-SD-SR consists of ~169M parameters, including UNet, encoder and decoder, and has a complexity of only ~142 GFLOPs. To maintain a high visual quality on such low compute budget, we introduce a number of training strategies: (i) A novel conditioning mechanism on the low resolution input, coined bidirectional conditioning, which tailors the SD model for the SR task. (ii) Joint training of the UNet and encoder, while decoupling the encodings of the HR and LR images and using a dedicated schedule. (iii) Finetuning the decoder using the UNet's output to directly tailor the decoder to the latents obtained at inference time. Edge-SD-SR runs efficiently on device, e.g. it can upscale a 128x128 patch to 512x512 in 38 msec while running on a Samsung S24 DSP, and of a 512x512 to 2048x2048 (requiring 25 model evaluations) in just ~1.1 sec. Furthermore, we show that Edge-SD-SR matches or even outperforms state-of-the-art SR approaches on the most established SR benchmarks.

Diffusion models are proficient at generating high-quality images. They are however effective only when operating at the resolution used during training. Inference at a scaled resolution leads to repetitive patterns and structural distortions. Retraining at higher resolutions quickly becomes prohibitive. Thus, methods enabling pre-existing diffusion models to operate at flexible test-time resolutions are highly desirable. Previous works suffer from frequent artifacts and often introduce large latency overheads. We propose two simple modules that combine to solve these issues. We introduce a Frequency Modulation (FM) module that leverages the Fourier domain to improve the global structure consistency, and an Attention Modulation (AM) module which improves the consistency of local texture patterns, a problem largely ignored in prior works. Our method, coined Fam diffusion, can seamlessly integrate into any latent diffusion model and requires no additional training. Extensive qualitative results highlight the effectiveness of our method in addressing structural and local artifacts, while quantitative results show state-of-the-art performance. Also, our method avoids redundant inference tricks for improved consistency such as patch-based or progressive generation, leading to negligible latency overheads.

Large Language Model (LLM) agents are rapidly improving to handle increasingly complex web-based tasks. Most of these agents rely on general-purpose, proprietary models like GPT-4 and focus on designing better prompts to improve their planning abilities. However, general-purpose LLMs are not specifically trained to understand specialized web contexts such as HTML, and they often struggle with long-horizon planning. We explore an alternative approach that fine-tunes open-source LLMs using production-scale workflow data collected from over 250 domains corresponding to 6 billion tokens. This simple yet effective approach shows substantial gains over prompting-based agents on existing benchmarks -- ScribeAgent achieves state-of-the-art direct generation performance on Mind2Web and improves the task success rate by 7.3% over the previous best text-only web agents on WebArena. We further perform detailed ablation studies on various fine-tuning design choices and provide insights into LLM selection, training recipes, context window optimization, and effect of dataset sizes.

In this paper, we present an end-to-end multi-source Entity Matching problem, which we call entity group matching, where the goal is to assign to the same group, records originating from multiple data sources but representing the same real-world entity. We focus on the effects of transitively matched records, i.e. the records connected by paths in the graph G = (V,E) whose nodes and edges represent the records and whether they are a match or not. We present a real-world instance of this problem, where the challenge is to match records of companies and financial securities originating from different data providers. We also introduce two new multi-source benchmark datasets that present similar matching challenges as real-world records. A distinctive characteristic of these records is that they are regularly updated following real-world events, but updates are not applied uniformly across data sources. This phenomenon makes the matching of certain groups of records only possible through the use of transitive information. In our experiments, we illustrate how considering transitively matched records is challenging since a limited amount of false positive pairwise match predictions can throw off the group assignment of large quantities of records. Thus, we propose GraLMatch, a method that can partially detect and remove false positive pairwise predictions through graph-based properties. Finally, we showcase how fine-tuning a Transformer-based model (DistilBERT) on a reduced number of labeled samples yields a better final entity group matching than training on more samples and/or incorporating fine-tuning optimizations, illustrating how precision becomes the deciding factor in the entity group matching of large volumes of records., Comment: 12 pages, 4 figures, accepted as research paper at EDBT 2025

Pre-trained Language Models (LMs) exhibit strong zero-shot and in-context learning capabilities; however, their behaviors are often difficult to control. By utilizing Reinforcement Learning from Human Feedback (RLHF), it is possible to fine-tune unsupervised LMs to follow instructions and produce outputs that reflect human preferences. Despite its benefits, RLHF has been shown to potentially harm a language model's reasoning capabilities and introduce artifacts such as hallucinations where the model may fabricate facts. To address this issue we introduce Direct Preference Heads (DPH), a fine-tuning framework that enables LMs to learn human preference signals through an auxiliary reward head without directly affecting the output distribution of the language modeling head. We perform a theoretical analysis of our objective function and find strong ties to Conservative Direct Preference Optimization (cDPO). Finally we evaluate our models on GLUE, RACE, and the GPT4All evaluation suite and demonstrate that our method produces models which achieve higher scores than those fine-tuned with Supervised Fine-Tuning (SFT) or Direct Preference Optimization (DPO) alone.

Nano-engineering crystalline materials can be used to tailor their thermal properties. By adding new nanoscale phonon scattering centers and controlling their size, one can effectively decrease the phonon mean free path and hence the thermal conductivity of a fully crystalline material. In this letter, we use the 3$\omega$ method in the temperature range of 100-300 K to experimentally report on the more than threefold reduction of the thermal conductivity of an epitaxially-grown crystalline germanium thin film with embedded polydispersed crystalline \ch{Ge3Mn5} nano-inclusions with diameters ranging from 5 to 25~nm. A detailed analysis of the structure of the thin film coupled with Monte Carlo simulations of phonon transport highlight the role of the nano-inclusions volume fraction in the reduction of the phononic contribution to the thermal conductivity, in particular its temperature dependence, leading to a phonon mean free path that is set by geometrical constraints., Comment: Applied Physics Letters, In press

Content and image generation consist in creating or generating data from noisy information by extracting specific features such as texture, edges, and other thin image structures. We are interested here in generative models, and two main problems are addressed. Firstly, the improvements of specific feature extraction while accounting at multiscale levels intrinsic geometric features; and secondly, the equivariance of the network to reduce its complexity and provide a geometric interpretability. To proceed, we propose a geometric generative model based on an equivariant partial differential equation (PDE) for group convolution neural networks (G-CNNs), so called PDE-G-CNNs, built on morphology operators and generative adversarial networks (GANs). Equivariant morphological PDE layers are composed of multiscale dilations and erosions formulated in Riemannian manifolds, while group symmetries are defined on a Lie group. We take advantage of the Lie group structure to properly integrate the equivariance in layers, and are able to use the Riemannian metric to solve the multiscale morphological operations. Each point of the Lie group is associated with a unique point in the manifold, which helps us derive a metric on the Riemannian manifold from a tensor field invariant under the Lie group so that the induced metric has the same symmetries. The proposed geometric morphological GAN (GM-GAN) is obtained by using the proposed morphological equivariant convolutions in PDE-G-CNNs to bring nonlinearity in classical CNNs. GM-GAN is evaluated on MNIST data and compared with GANs. Preliminary results show that GM-GAN model outperforms classical GAN.

Blockchain technology, a foundational distributed ledger system, enables secure and transparent multi-party transactions. Despite its advantages, blockchain networks are susceptible to anomalies and frauds, posing significant risks to their integrity and security. This paper offers a detailed examination of blockchain's key definitions and properties, alongside a thorough analysis of the various anomalies and frauds that undermine these networks. It describes an array of detection and prevention strategies, encompassing statistical and machine learning methods, game-theoretic solutions, digital forensics, reputation-based systems, and comprehensive risk assessment techniques. Through case studies, we explore practical applications of anomaly and fraud detection in blockchain networks, extracting valuable insights and implications for both current practice and future research. Moreover, we spotlight emerging trends and challenges within the field, proposing directions for future investigation and technological development. Aimed at both practitioners and researchers, this paper seeks to provide a technical, in-depth overview of anomaly and fraud detection within blockchain networks, marking a significant step forward in the search for enhanced network security and reliability.

In this paper, we introduce YONOS-SR, a novel stable diffusion-based approach for image super-resolution that yields state-of-the-art results using only a single DDIM step. We propose a novel scale distillation approach to train our SR model. Instead of directly training our SR model on the scale factor of interest, we start by training a teacher model on a smaller magnification scale, thereby making the SR problem simpler for the teacher. We then train a student model for a higher magnification scale, using the predictions of the teacher as a target during the training. This process is repeated iteratively until we reach the target scale factor of the final model. The rationale behind our scale distillation is that the teacher aids the student diffusion model training by i) providing a target adapted to the current noise level rather than using the same target coming from ground truth data for all noise levels and ii) providing an accurate target as the teacher has a simpler task to solve. We empirically show that the distilled model significantly outperforms the model trained for high scales directly, specifically with few steps during inference. Having a strong diffusion model that requires only one step allows us to freeze the U-Net and fine-tune the decoder on top of it. We show that the combination of spatially distilled U-Net and fine-tuned decoder outperforms state-of-the-art methods requiring 200 steps with only one single step.

In this paper, we focus on task-specific question answering (QA). To this end, we introduce a method for generating exhaustive and high-quality training data, which allows us to train compact (e.g., run on a mobile device), task-specific QA models that are competitive against GPT variants. The key technological enabler is a novel mechanism for automatic question-answer generation from procedural text which can ingest large amounts of textual instructions and produce exhaustive in-domain QA training data. While current QA data generation methods can produce well-formed and varied data, their non-exhaustive nature is sub-optimal for training a QA model. In contrast, we leverage the highly structured aspect of procedural text and represent each step and the overall flow of the procedure as graphs. We then condition on graph nodes to automatically generate QA pairs in an exhaustive and controllable manner. Comprehensive evaluations of our method show that: 1) small models trained with our data achieve excellent performance on the target QA task, even exceeding that of GPT3 and ChatGPT despite being several orders of magnitude smaller. 2) semantic coverage is the key indicator for downstream QA performance. Crucially, while large language models excel at syntactic diversity, this does not necessarily result in improvements on the end QA model. In contrast, the higher semantic coverage provided by our method is critical for QA performance., Comment: Accepted to EACL 2024 as long paper. 25 pages including appendix

We consider the accuracy of an approximate posterior distribution in nonparametric regression problems by combining posterior distributions computed on subsets of the data defined by the locations of the independent variables. We show that this approximate posterior retains the rate of recovery of the full data posterior distribution, where the rate of recovery adapts to the smoothness of the true regression function. As particular examples we consider Gaussian process priors based on integrated Brownian motion and the Mat\'ern kernel augmented with a prior on the length scale. Besides theoretical guarantees we present a numerical study of the methods both on synthetic and real world data. We also propose a new aggregation technique, which numerically outperforms previous approaches., Comment: 20 pages, 4 figures

This paper introduces a novel Parameter-Efficient Fine-Tuning (PEFT) framework for multi-modal, multi-task transfer learning with pre-trained language models. PEFT techniques such as LoRA, BitFit and IA3 have demonstrated comparable performance to full fine-tuning of pre-trained models for specific downstream tasks, all while demanding significantly fewer trainable parameters and reduced GPU memory consumption. However, in the context of multi-modal fine-tuning, the need for architectural modifications or full fine-tuning often becomes apparent. To address this we propose Context-PEFT, which learns different groups of adaptor parameters based on the token's domain. This approach enables LoRA-like weight injection without requiring additional architectural changes. Our method is evaluated on the COCO captioning task, where it outperforms full fine-tuning under similar data constraints while simultaneously offering a substantially more parameter-efficient and computationally economical solution.

Machine learning and deep learning have become increasingly prevalent in financial prediction and forecasting tasks, offering advantages such as enhanced customer experience, democratising financial services, improving consumer protection, and enhancing risk management. However, these complex models often lack transparency and interpretability, making them challenging to use in sensitive domains like finance. This has led to the rise of eXplainable Artificial Intelligence (XAI) methods aimed at creating models that are easily understood by humans. Classical XAI methods, such as LIME and SHAP, have been developed to provide explanations for complex models. While these methods have made significant contributions, they also have limitations, including computational complexity, inherent model bias, sensitivity to data sampling, and challenges in dealing with feature dependence. In this context, this paper explores good practices for deploying explainability in AI-based systems for finance, emphasising the importance of data quality, audience-specific methods, consideration of data properties, and the stability of explanations. These practices aim to address the unique challenges and requirements of the financial industry and guide the development of effective XAI tools., Comment: 11 pages, 1 figure

We study the problem of future step anticipation in procedural videos. Given a video of an ongoing procedural activity, we predict a plausible next procedure step described in rich natural language. While most previous work focus on the problem of data scarcity in procedural video datasets, another core challenge of future anticipation is how to account for multiple plausible future realizations in natural settings. This problem has been largely overlooked in previous work. To address this challenge, we frame future step prediction as modelling the distribution of all possible candidates for the next step. Specifically, we design a generative model that takes a series of video clips as input, and generates multiple plausible and diverse candidates (in natural language) for the next step. Following previous work, we side-step the video annotation scarcity by pretraining our model on a large text-based corpus of procedural activities, and then transfer the model to the video domain. Our experiments, both in textual and video domains, show that our model captures diversity in the next step prediction and generates multiple plausible future predictions. Moreover, our model establishes new state-of-the-art results on YouCookII, where it outperforms existing baselines on the next step anticipation. Finally, we also show that our model can successfully transfer from text to the video domain zero-shot, ie, without fine-tuning or adaptation, and produces good-quality future step predictions from video., Comment: published at ICCV 2023

For $N\geq 4$, we let $\Omega$ be a bounded domain of $\mathbb{R}^N$ and $\Gamma$ be a closed curve contained in $\Omega$. We study existence of positive solutions $u \in H^1_0\left(\Omega\right)$ to the equation \begin{equation}\label{Atusi} -\Delta u+hu=\lambda\rho^{-s_1}_\Gamma u^{2^*_{s_1}-1}+\rho^{-s_2}_\Gamma u^{2^*_{s_2}-1} \qquad \textrm{ in } \Omega \end{equation} where $h$ is a continuous function and $\rho_\Gamma$ is the distance function to $\Gamma$. We prove the existence of a mountain pass solution for this Euler-Lagrange equation depending on the local geometry of the curve and the potential $h$., Comment: arXiv admin note: text overlap with arXiv:1702.02202

We let $\Omega$ be a bounded domain of $\mathbb{R}^3$ and $\Gamma$ be a closed curve contained in $\Omega$. We study existence of positive solutions $u \in H^1_0\left(\Omega\right)$ to the equation $$ -\Delta u+hu=\lambda\rho^{-s_1}_\Gamma u^{5-2s_1}+\rho^{-s_2}_\Gamma u^{5-2s_2} \qquad \textrm{ in } \Omega $$ where $h$ is a continuous function and $\rho_\Gamma$ is the distance function to $\Gamma$. We prove existence of solutions depending on the regular part of the Green function of linear operator. We prove the existence of positive mountain pass solutions for this Euler-Lagrange equation depending on the mass which is the regular part of the Green function of the linear operator $-\Delta+h$., Comment: arXiv admin note: substantial text overlap with arXiv:1702.02202