Giorgio Giannone

Matteo Manica, Jannis Born, Joris Cadow et al. · mit

With the growing availability of data within various scientific domains, generative models hold enormous potential to accelerate scientific discovery. They harness powerful representations learned from datasets to speed up the formulation of novel hypotheses with the potential to impact material discovery broadly. We present the Generative Toolkit for Scientific Discovery (GT4SD). This extensible open-source library enables scientists, developers, and researchers to train and use state-of-the-art generative models to accelerate scientific discovery focused on material design.

Lyle Regenwetter, Giorgio Giannone, Akash Srivastava et al. · mit

Generative models have recently achieved remarkable success and widespread adoption in society, yet they often struggle to generate realistic and accurate outputs. This challenge extends beyond language and vision into fields like engineering design, where safety-critical engineering standards and non-negotiable physical laws tightly constrain what outputs are considered acceptable. In this work, we introduce a novel training method to guide a generative model toward constraint-satisfying outputs using `negative data' -- examples of what to avoid. Our negative-data generative model (NDGM) formulation easily outperforms classic models, generating 1/6 as many constraint-violating samples using 1/8 as much data in certain problems. It also consistently outperforms other baselines, achieving a balance between constraint satisfaction and distributional similarity that is unsurpassed by any other model in 12 of the 14 problems tested. This widespread superiority is rigorously demonstrated across numerous synthetic tests and real engineering problems, such as ship hull synthesis with hydrodynamic constraints and vehicle design with impact safety constraints. Our benchmarks showcase both the best-in-class performance of our new NDGM formulation and the overall dominance of NDGMs versus classic generative models. We publicly release the code and benchmarks at https://github.com/Lyleregenwetter/NDGMs.

Dimitrios Christofidellis, Giorgio Giannone, Jannis Born et al. · mit

The recent advances in neural language models have also been successfully applied to the field of chemistry, offering generative solutions for classical problems in molecular design and synthesis planning. These new methods have the potential to fuel a new era of data-driven automation in scientific discovery. However, specialized models are still typically required for each task, leading to the need for problem-specific fine-tuning and neglecting task interrelations. The main obstacle in this field is the lack of a unified representation between natural language and chemical representations, complicating and limiting human-machine interaction. Here, we propose the first multi-domain, multi-task language model that can solve a wide range of tasks in both the chemical and natural language domains. Our model can handle chemical and natural language concurrently, without requiring expensive pre-training on single domains or task-specific models. Interestingly, sharing weights across domains remarkably improves our model when benchmarked against state-of-the-art baselines on single-domain and cross-domain tasks. In particular, sharing information across domains and tasks gives rise to large improvements in cross-domain tasks, the magnitude of which increase with scale, as measured by more than a dozen of relevant metrics. Our work suggests that such models can robustly and efficiently accelerate discovery in physical sciences by superseding problem-specific fine-tuning and enhancing human-model interactions.

Giorgio Giannone, Didrik Nielsen, Ole Winther · mit

Denoising diffusion probabilistic models (DDPM) are powerful hierarchical latent variable models with remarkable sample generation quality and training stability. These properties can be attributed to parameter sharing in the generative hierarchy, as well as a parameter-free diffusion-based inference procedure. In this paper, we present Few-Shot Diffusion Models (FSDM), a framework for few-shot generation leveraging conditional DDPMs. FSDMs are trained to adapt the generative process conditioned on a small set of images from a given class by aggregating image patch information using a set-based Vision Transformer (ViT). At test time, the model is able to generate samples from previously unseen classes conditioned on as few as 5 samples from that class. We empirically show that FSDM can perform few-shot generation and transfer to new datasets. We benchmark variants of our method on complex vision datasets for few-shot learning and compare to unconditional and conditional DDPM baselines. Additionally, we show how conditioning the model on patch-based input set information improves training convergence.

Harry Jake Cunningham, Giorgio Giannone, Mingtian Zhang et al. · mit

Global convolutions have shown increasing promise as powerful general-purpose sequence models. However, training long convolutions is challenging, and kernel parameterizations must be able to learn long-range dependencies without overfitting. This work introduces reparameterized multi-resolution convolutions ($\texttt{MRConv}$), a novel approach to parameterizing global convolutional kernels for long-sequence modelling. By leveraging multi-resolution convolutions, incorporating structural reparameterization and introducing learnable kernel decay, $\texttt{MRConv}$ learns expressive long-range kernels that perform well across various data modalities. Our experiments demonstrate state-of-the-art performance on the Long Range Arena, Sequential CIFAR, and Speech Commands tasks among convolution models and linear-time transformers. Moreover, we report improved performance on ImageNet classification by replacing 2D convolutions with 1D $\texttt{MRConv}$ layers.

Giorgio Giannone, Faez Ahmed · mit

Topology Optimization seeks to find the best design that satisfies a set of constraints while maximizing system performance. Traditional iterative optimization methods like SIMP can be computationally expensive and get stuck in local minima, limiting their applicability to complex or large-scale problems. Learning-based approaches have been developed to accelerate the topology optimization process, but these methods can generate designs with floating material and low performance when challenged with out-of-distribution constraint configurations. Recently, deep generative models, such as Generative Adversarial Networks and Diffusion Models, conditioned on constraints and physics fields have shown promise, but they require extensive pre-processing and surrogate models for improving performance. To address these issues, we propose a Generative Optimization method that integrates classic optimization like SIMP as a refining mechanism for the topology generated by a deep generative model. We also remove the need for conditioning on physical fields using a computationally inexpensive approximation inspired by classic ODE solutions and reduce the number of steps needed to generate a feasible and performant topology. Our method allows us to efficiently generate good topologies and explicitly guide them to regions with high manufacturability and high performance, without the need for external auxiliary models or additional labeled data. We believe that our method can lead to significant advancements in the design and optimization of structures in engineering applications, and can be applied to a broader spectrum of performance-aware engineering design problems.

Cyril Picard, Kristen M. Edwards, Anna C. Doris et al. · mit

Engineering design is undergoing a transformative shift with the advent of AI, marking a new era in how we approach product, system, and service planning. Large language models have demonstrated impressive capabilities in enabling this shift. Yet, with text as their only input modality, they cannot leverage the large body of visual artifacts that engineers have used for centuries and are accustomed to. This gap is addressed with the release of multimodal vision-language models (VLMs), such as GPT-4V, enabling AI to impact many more types of tasks. Our work presents a comprehensive evaluation of VLMs across a spectrum of engineering design tasks, categorized into four main areas: Conceptual Design, System-Level and Detailed Design, Manufacturing and Inspection, and Engineering Education Tasks. Specifically in this paper, we assess the capabilities of two VLMs, GPT-4V and LLaVA 1.6 34B, in design tasks such as sketch similarity analysis, CAD generation, topology optimization, manufacturability assessment, and engineering textbook problems. Through this structured evaluation, we not only explore VLMs' proficiency in handling complex design challenges but also identify their limitations in complex engineering design applications. Our research establishes a foundation for future assessments of vision language models. It also contributes a set of benchmark testing datasets, with more than 1000 queries, for ongoing advancements and applications in this field.

Giorgio Giannone, Serhii Havrylov, Jordan Massiah et al. · mit

Advances in deep learning theory have revealed how average generalization relies on superficial patterns in data. The consequences are brittle models with poor performance with shift in group distribution at test time. When group annotation is available, we can use robust optimization tools to tackle the problem. However, identification and annotation are time-consuming, especially on large datasets. A recent line of work leverages self-supervision and oversampling to improve generalization on minority groups without group annotation. We propose to unify and generalize these approaches using a class-conditional variant of mixup tailored for worst-group generalization. Our approach, Just Mix Once (JM1), interpolates samples during learning, augmenting the training distribution with a continuous mixture of groups. JM1 is domain agnostic and computationally efficient, can be used with any level of group annotation, and performs on par or better than the state-of-the-art on worst-group generalization. Additionally, we provide a simple explanation of why JM1 works.

Giorgio Giannone, Anna Clare Doris, Amin Heyrani Nobari et al.

Generating executable CAD programs from images requires alignment between visual geometry and symbolic program representations, a capability that current methods fail to learn reliably as design complexity increases. Existing fine-tuning approaches rely on either limited supervised datasets or expensive post-training pipelines, resulting in brittle systems that restrict progress in generative CAD design. We argue that the primary bottleneck lies not in model or algorithmic capacity, but in the scarcity of diverse training examples that align visual geometry with program syntax. This limitation is especially acute because the collection of diverse and verified engineering datasets is both expensive and difficult to scale, constraining the development of robust generative CAD models. We introduce Geometric Inference Feedback Tuning (GIFT), a data augmentation framework that leverages geometric feedback to turn test-time compute into a bootstrapped set of high-quality training samples. GIFT combines two mechanisms: Soft-Rejection Sampling (GIFT-REJECT), which retains diverse high-fidelity programs beyond exact ground-truth matches, and Failure-Driven Augmentation (GIFT-FAIL), which converts near-miss predictions into synthetic training examples that improve robustness on challenging geometries. By amortizing inference-time search into the model parameters, GIFT captures the benefits of test-time scaling while reducing inference compute by 80%. It improves mean IoU by 12% over a strong supervised baseline and remains competitive with more complex multimodal systems, without requiring additional human annotation or specialized architectures.

Amin Heyrani Nobari, Giorgio Giannone, Lyle Regenwetter et al. · mit

Topology optimization is a critical task in engineering design, where the goal is to optimally distribute material in a given space for maximum performance. We introduce Neural Implicit Topology Optimization (NITO), a novel approach to accelerate topology optimization problems using deep learning. NITO stands out as one of the first frameworks to offer a resolution-free and domain-agnostic solution in deep learning-based topology optimization. NITO synthesizes structures with up to seven times better structural efficiency compared to SOTA diffusion models and does so in a tenth of the time. In the NITO framework, we introduce a novel method, the Boundary Point Order-Invariant MLP (BPOM), to represent boundary conditions in a sparse and domain-agnostic manner, moving away from expensive simulation-based approaches. Crucially, NITO circumvents the domain and resolution limitations that restrict Convolutional Neural Network (CNN) models to a structured domain of fixed size -- limitations that hinder the widespread adoption of CNNs in engineering applications. This generalizability allows a single NITO model to train and generate solutions in countless domains, eliminating the need for numerous domain-specific CNNs and their extensive datasets. Despite its generalizability, NITO outperforms SOTA models even in specialized tasks, is an order of magnitude smaller, and is practically trainable at high resolutions that would be restrictive for CNNs. This combination of versatility, efficiency, and performance underlines NITO's potential to transform the landscape of engineering design optimization problems through implicit fields.

Giorgio Giannone, Guangxuan Xu, Nikhil Shivakumar Nayak et al. · mit

Inference-Time Scaling (ITS) improves language models by allocating more computation at generation time. Particle Filtering (PF) has emerged as a strong ITS method for complex mathematical reasoning tasks, but it is vulnerable when guided by process reward models, which often assign overconfident scores early in the reasoning process. This causes PF to suffer from premature exploitation: it myopically commits to locally promising trajectories, prunes potentially correct hypotheses, and converges to suboptimal solutions. This failure mode, known as particle impoverishment, is especially severe under constrained computational budgets. To address this, we analyze the problem and identify two root causes: a lack of diversity in the particle set due to overconfident resampling and consequent inability to assess the potential of a reasoning path. We introduce Entropic Particle Filtering (ePF), an algorithm that integrates two new techniques to solve these issues. The first technique, Entropic Annealing (EA), directly mitigates particle impoverishment by monitoring search diversity via entropy; when diversity drops, it intervenes by dynamically annealing the resampling distribution to preserve exploration. The second, an enhancement called Look-ahead Modulation (LaM), adds a predictive guide to evaluate a state's potential based on its successors. On several challenging math benchmarks, ePF significantly outperforms strong baselines and achieves up to a 50 % relative improvement in task reward. Together, these methods improve PF's resilience by balancing the exploration of diverse solution spaces with the exploitation of high-reward regions, ultimately leading to higher-quality solutions.

Giorgio Giannone, Ruoteng Li, Qianli Feng et al.

Vision-language models (VLMs) have demonstrated remarkable potential in integrating visual and linguistic information, but their performance is often constrained by the need for extensive, high-quality image-text training data. Curation of these image-text pairs is both time-consuming and computationally expensive. To address this challenge, we introduce SVP (Sampling-based Visual Projection), a novel framework that enhances vision-language alignment without relying on manually curated text-image pairs or preference annotation. SVP leverages a small set of manually selected images, self-captioning and a pre-trained grounding model as a feedback mechanism to elicit latent information in VLMs. We evaluate our approach across six key areas: captioning, referring, visual question answering, multitasking, hallucination control, and object recall. Results demonstrate significant improvements, including a 14 % average improvement in captioning tasks, up to 12 % increase in object recall, and significantly reduced hallucinations, while maintaining question-answering capabilities. Using SVP, a small VLM achieves hallucination reductions similar to a model five times larger, while a VLM with initially poor referring capabilities more than doubles its performance, approaching parity with a model twice its size.

Giorgio Giannone, Akash Srivastava, Ole Winther et al.

Generative models have had a profound impact on vision and language, paving the way for a new era of multimodal generative applications. While these successes have inspired researchers to explore using generative models in science and engineering to accelerate the design process and reduce the reliance on iterative optimization, challenges remain. Specifically, engineering optimization methods based on physics still outperform generative models when dealing with constrained environments where data is scarce and precision is paramount. To address these challenges, we introduce Diffusion Optimization Models (DOM) and Trajectory Alignment (TA), a learning framework that demonstrates the efficacy of aligning the sampling trajectory of diffusion models with the optimization trajectory derived from traditional physics-based methods. This alignment ensures that the sampling process remains grounded in the underlying physical principles. Our method allows for generating feasible and high-performance designs in as few as two steps without the need for expensive preprocessing, external surrogate models, or additional labeled data. We apply our framework to structural topology optimization, a fundamental problem in mechanical design, evaluating its performance on in- and out-of-distribution configurations. Our results demonstrate that TA outperforms state-of-the-art deep generative models on in-distribution configurations and halves the inference computational cost. When coupled with a few steps of optimization, it also improves manufacturability for out-of-distribution conditions. By significantly improving performance and inference efficiency, DOM enables us to generate high-quality designs in just a few steps and guide them toward regions of high performance and manufacturability, paving the way for the widespread application of generative models in large-scale data-driven design.

Giorgio Giannone, Ole Winther

A few-shot generative model should be able to generate data from a novel distribution by only observing a limited set of examples. In few-shot learning the model is trained on data from many sets from distributions sharing some underlying properties such as sets of characters from different alphabets or objects from different categories. We extend current latent variable models for sets to a fully hierarchical approach with an attention-based point to set-level aggregation and call our method SCHA-VAE for Set-Context-Hierarchical-Aggregation Variational Autoencoder. We explore likelihood-based model comparison, iterative data sampling, and adaptation-free out-of-distribution generalization. Our results show that the hierarchical formulation better captures the intrinsic variability within the sets in the small data regime. This work generalizes deep latent variable approaches to few-shot learning, taking a step toward large-scale few-shot generation with a formulation that readily works with current state-of-the-art deep generative models.

Giorgio Giannone, Asha Anoosheh, Alessio Quaglino et al.

Event-based cameras are novel, efficient sensors inspired by the human vision system, generating an asynchronous, pixel-wise stream of data. Learning from such data is generally performed through heavy preprocessing and event integration into images. This requires buffering of possibly long sequences and can limit the response time of the inference system. In this work, we instead propose to directly use events from a DVS camera, a stream of intensity changes and their spatial coordinates. This sequence is used as the input for a novel \emph{asynchronous} RNN-like architecture, the Input-filtering Neural ODEs (INODE). This is inspired by the dynamical systems and filtering literature. INODE is an extension of Neural ODEs (NODE) that allows for input signals to be continuously fed to the network, like in filtering. The approach naturally handles batches of time series with irregular time-stamps by implementing a batch forward Euler solver. INODE is trained like a standard RNN, it learns to discriminate short event sequences and to perform event-by-event online inference. We demonstrate our approach on a series of classification tasks, comparing against a set of LSTM baselines. We show that, independently of the camera resolution, INODE can outperform the baselines by a large margin on the ASL task and it's on par with a much larger LSTM for the NCALTECH task. Finally, we show that INODE is accurate even when provided with very few events.

Giorgio Giannone, Saeed Saremi, Jonathan Masci et al.

We extend the framework of variational autoencoders to represent transformations explicitly in the latent space. In the family of hierarchical graphical models that emerges, the latent space is populated by higher order objects that are inferred jointly with the latent representations they act on. To explicitly demonstrate the effect of these higher order objects, we show that the inferred latent transformations reflect interpretable properties in the observation space. Furthermore, the model is structured in such a way that in the absence of transformations, we can run inference and obtain generative capabilities comparable with standard variational autoencoders. Finally, utilizing the trained encoder, we outperform the baselines by a wide margin on a challenging out-of-distribution classification task.

Giorgio Giannone, Boris Chidlovskii

We propose a new deep learning architecture for the tasks of semantic segmentation and depth prediction from RGB-D images. We revise the state of art based on the RGB and depth feature fusion, where both modalities are assumed to be available at train and test time. We propose a new architecture where the feature fusion is replaced with a common deep representation. Combined with an encoder-decoder type of the network, the architecture can jointly learn models for semantic segmentation and depth estimation based on their common representation. This representation, inspired by multi-view learning, offers several important advantages, such as using one modality available at test time to reconstruct the missing modality. In the RGB-D case, this enables the cross-modality scenarios, such as using depth data for semantically segmentation and the RGB images for depth estimation. We demonstrate the effectiveness of the proposed network on two publicly available RGB-D datasets. The experimental results show that the proposed method works well in both semantic segmentation and depth estimation tasks.