Search Results for author: Giorgio Giannone

Found 15 papers, 6 papers with code

Feedback-Driven Vision-Language Alignment with Minimal Human Supervision

no code implementations8 Jan 2025 Giorgio Giannone, Ruoteng Li, Qianli Feng, Evgeny Perevodchikov, Rui Chen, Aleix Martinez

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.

Hallucination Question Answering +1

Be More Specific: Evaluating Object-centric Realism in Synthetic Images

no code implementations CVPR 2025 Anqi Liang, Ciprian Corneanu, Qianli Feng, Giorgio Giannone, Aleix Martinez

In this work, we define a new standard for assessing object-centric realism that follows a shape-texture breakdown and proposes the first object-centric realism evaluation dataset for synthetic images.

Object

Reparameterized Multi-Resolution Convolutions for Long Sequence Modelling

no code implementations18 Aug 2024 Harry Jake Cunningham, Giorgio Giannone, Mingtian Zhang, Marc Peter Deisenroth

Global convolutions have shown increasing promise as powerful general-purpose sequence models.

NITO: Neural Implicit Fields for Resolution-free Topology Optimization

1 code implementation7 Feb 2024 Amin Heyrani Nobari, Giorgio Giannone, Lyle Regenwetter, Faez Ahmed

We introduce Neural Implicit Topology Optimization (NITO), a novel approach to accelerate topology optimization problems using deep learning.

Constraining Generative Models for Engineering Design with Negative Data

1 code implementation27 Jun 2023 Lyle Regenwetter, Giorgio Giannone, Akash Srivastava, Dan Gutfreund, Faez Ahmed

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.

Diversity valid

Diffusing the Optimal Topology: A Generative Optimization Approach

no code implementations17 Mar 2023 Giorgio Giannone, Faez Ahmed

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.

Unifying Molecular and Textual Representations via Multi-task Language Modelling

1 code implementation29 Jan 2023 Dimitrios Christofidellis, Giorgio Giannone, Jannis Born, Ole Winther, Teodoro Laino, Matteo Manica

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.

Language Modelling Molecule Captioning +3

Just Mix Once: Worst-group Generalization by Group Interpolation

no code implementations21 Oct 2022 Giorgio Giannone, Serhii Havrylov, Jordan Massiah, Emine Yilmaz, Yunlong Jiao

Advances in deep learning theory have revealed how average generalization relies on superficial patterns in data.

Learning Theory

Few-Shot Diffusion Models

1 code implementation30 May 2022 Giorgio Giannone, Didrik Nielsen, Ole Winther

At test time, the model is able to generate samples from previously unseen classes conditioned on as few as 5 samples from that class.

Denoising Few-Shot Learning

SCHA-VAE: Hierarchical Context Aggregation for Few-Shot Generation

1 code implementation23 Oct 2021 Giorgio Giannone, Ole Winther

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.

Few-Shot Learning Out-of-Distribution Generalization

No Representation without Transformation

no code implementations9 Dec 2019 Giorgio Giannone, Saeed Saremi, Jonathan Masci, Christian Osendorfer

To explicitly demonstrate the effect of these higher order objects, we show that the inferred latent transformations reflect interpretable properties in the observation space.

Learning Common Representation from RGB and Depth Images

no code implementations17 Dec 2018 Giorgio Giannone, Boris Chidlovskii

We propose a new deep learning architecture for the tasks of semantic segmentation and depth prediction from RGB-D images.

Decoder Depth Estimation +4

Cannot find the paper you are looking for? You can Submit a new open access paper.