Search Results for author:
Found 80 papers, 44 papers with code
Towards Precise Completion of Deformable Shapes
Here, we claim that observing part of an object which was previously acquired as a whole, one could deal with both partial matching and shape completion in a holistic manner.
Video Unlearning via Low-Rank Refusal Vector
This isolates the target concept while minimizing collateral unlearning of other semantics, thus preserving the visual quality of the generated video.
Implicit Inversion turns CLIP into a Decoder
CLIP is a discriminative model trained to align images and text in a shared embedding space.
Update Your Transformer to the Latest Release: Re-Basin of Task Vectors
In this work, we investigate how to transfer fine-tuning to a new checkpoint without having to re-train, in a data-free manner.
Navigating the Latent Space Dynamics of Neural Models
Neural networks transform high-dimensional data into compact, structured representations, often modeled as elements of a lower dimensional latent space.
Mergenetic: a Simple Evolutionary Model Merging Library
Model merging allows combining the capabilities of existing models into a new one - post hoc, without additional training.
Model-based Metric 3D Shape and Motion Reconstruction of Wild Bottlenose Dolphins in Drone-Shot Videos
We address the problem of estimating the metric 3D shape and motion of wild dolphins from monocular video, with the aim of assessing their body condition.
MASS: MoErging through Adaptive Subspace Selection
Model merging has recently emerged as a lightweight alternative to ensembling, combining multiple fine-tuned models into a single set of parameters with no additional training overhead.
LoopGen: Training-Free Loopable Music Generation
We address this gap by modifying a non-autoregressive model (MAGNeT) to generate tokens in a circular pattern, letting the model attend to the beginning of the audio when creating its ending.
Escaping Plato's Cave: Towards the Alignment of 3D and Text Latent Spaces
In this work, we investigate the possibility of a posteriori alignment of representations obtained from uni-modal 3D encoders compared to text-based feature spaces.
MERGE$^3$: Efficient Evolutionary Merging on Consumer-grade GPUs
Evolutionary model merging enables the creation of high-performing multi-task models but remains computationally prohibitive for consumer hardware.
Humanity's Last Exam
However, benchmarks are not keeping pace in difficulty: LLMs now achieve over 90\% accuracy on popular benchmarks like MMLU, limiting informed measurement of state-of-the-art LLM capabilities.
Ranked #4 on
Humanity's Last Exam
on Humanity's Last Exam
Task Singular Vectors: Reducing Task Interference in Model Merging
In this paper, we study task vectors at the layer level, focusing on task layer matrices and their singular value decomposition.
ATM: Improving Model Merging by Alternating Tuning and Merging
Given this parallel between task vectors and gradients, we propose viewing model merging as a single step in an iterative process that alternates between tuning and merging (ATM).
ResiDual Transformer Alignment with Spectral Decomposition
When examined through the lens of their residual streams, a puzzling property emerges in transformer networks: residual contributions (e. g., attention heads) sometimes specialize in specific tasks or input attributes.
Detecting and Approximating Redundant Computational Blocks in Neural Networks
In this paper, we investigate the emergence of these internal similarities across different layers in diverse neural architectures, showing that similarity patterns emerge independently of the datataset used.
Preserving Privacy in Large Language Models: A Survey on Current Threats and Solutions
Large Language Models (LLMs) represent a significant advancement in artificial intelligence, finding applications across various domains.
Latent Space Translation via Inverse Relative Projection
By formalizing the invertibility of angle-preserving relative representations and assuming the scale invariance of decoder modules in neural models, we can effectively use the relative space as an intermediary, independently projecting onto and from other semantically similar spaces.
Latent Functional Maps: a spectral framework for representation alignment
Neural models learn data representations that lie on low-dimensional manifolds, yet modeling the relation between these representational spaces is an ongoing challenge.
$C^2M^3$: Cycle-Consistent Multi-Model Merging
In this paper, we present a novel data-free method for merging neural networks in weight space.
Implicit-ARAP: Efficient Handle-Guided Deformation of High-Resolution Meshes and Neural Fields via Local Patch Meshing
Then, we apply this representation in the setting of handle-guided deformation: we introduce two distinct pipelines, which make use of 3D neural fields to compute As-Rigid-As-Possible deformations of both high-resolution meshes and neural fields under a given set of constraints.
Naturalistic Music Decoding from EEG Data via Latent Diffusion Models
In this article, we explore the potential of using latent diffusion models, a family of powerful generative models, for the task of reconstructing naturalistic music from electroencephalogram (EEG) recordings.
COCOLA: Coherence-Oriented Contrastive Learning of Musical Audio Representations
We present COCOLA (Coherence-Oriented Contrastive Learning for Audio), a contrastive learning method for musical audio representations that captures the harmonic and rhythmic coherence between samples.
Zero-Shot Stitching in Reinforcement Learning using Relative Representations
We build upon the recent relative representations framework and adapt it for Visual RL.
Generalized Multi-Source Inference for Text Conditioned Music Diffusion Models
Multi-Source Diffusion Models (MSDM) allow for compositional musical generation tasks: generating a set of coherent sources, creating accompaniments, and performing source separation.
GSEdit: Efficient Text-Guided Editing of 3D Objects via Gaussian Splatting
We present GSEdit, a pipeline for text-guided 3D object editing based on Gaussian Splatting models.
Zero-Shot Duet Singing Voices Separation with Diffusion Models
In this paper, we examine this problem in the context of duet singing voices separation, and propose a method to enforce the coherency of singer identity by splitting the mixture into overlapping segments and performing posterior sampling in an auto-regressive manner, conditioning on the previous segment.
From Charts to Atlas: Merging Latent Spaces into One
Models trained on semantically related datasets and tasks exhibit comparable inter-sample relations within their latent spaces.
Latent Space Translation via Semantic Alignment
While different neural models often exhibit latent spaces that are alike when exposed to semantically related data, this intrinsic similarity is not always immediately discernible.
SyncFusion: Multimodal Onset-synchronized Video-to-Audio Foley Synthesis
Sound design involves creatively selecting, recording, and editing sound effects for various media like cinema, video games, and virtual/augmented reality.
From Bricks to Bridges: Product of Invariances to Enhance Latent Space Communication
It has been observed that representations learned by distinct neural networks conceal structural similarities when the models are trained under similar inductive biases.
Camoscio: an Italian Instruction-tuned LLaMA
In recent years Large Language Models (LLMs) have increased the state of the art on several natural language processing tasks.
Vector Quantile Regression on Manifolds
Quantile regression (QR) is a statistical tool for distribution-free estimation of conditional quantiles of a target variable given explanatory features.
Fauno: The Italian Large Language Model that will leave you senza parole!
This paper presents Fauno, the first and largest open-source Italian conversational Large Language Model (LLM).
Geometric Epitope and Paratope Prediction
Antibody-antigen interactions play a crucial role in identifying and neutralizing harmful foreign molecules.
Accelerating Transformer Inference for Translation via Parallel Decoding
We propose to reframe the standard greedy autoregressive decoding of MT with a parallel formulation leveraging Jacobi and Gauss-Seidel fixed-point iteration methods for fast inference.
Multimodal Neural Databases
The rise in loosely-structured data available through text, images, and other modalities has called for new ways of querying them.
Leveraging sparse and shared feature activations for disentangled representation learning
Recovering the latent factors of variation of high dimensional data has so far focused on simple synthetic settings.
Fluid Dynamics Network: Topology-Agnostic 4D Reconstruction via Fluid Dynamics Priors
Representing 3D surfaces as level sets of continuous functions over $\mathbb{R}^3$ is the common denominator of neural implicit representations, which recently enabled remarkable progress in geometric deep learning and computer vision tasks.
Bootstrapping Parallel Anchors for Relative Representations
The use of relative representations for latent embeddings has shown potential in enabling latent space communication and zero-shot model stitching across a wide range of applications.
Multi-Source Diffusion Models for Simultaneous Music Generation and Separation
In this work, we define a diffusion-based generative model capable of both music synthesis and source separation by learning the score of the joint probability density of sources sharing a context.
Latent Spectral Regularization for Continual Learning
While biological intelligence grows organically as new knowledge is gathered throughout life, Artificial Neural Networks forget catastrophically whenever they face a changing training data distribution.
Latent Autoregressive Source Separation
Autoregressive models have achieved impressive results over a wide range of domains in terms of generation quality and downstream task performance.
ASIF: Coupled Data Turns Unimodal Models to Multimodal Without Training
CLIP proved that aligning visual and language spaces is key to solving many vision tasks without explicit training, but required to train image and text encoders from scratch on a huge dataset.
Relative representations enable zero-shot latent space communication
Neural networks embed the geometric structure of a data manifold lying in a high-dimensional space into latent representations.
Sparse Vicious Attacks on Graph Neural Networks
In this work, we focus on a specific, white-box attack to GNN-based link prediction models, where a malicious node aims to appear in the list of recommended nodes for a given target victim.
KiloNeuS: A Versatile Neural Implicit Surface Representation for Real-Time Rendering
NeRF-based techniques fit wide and deep multi-layer perceptrons (MLPs) to a continuous radiance field that can be rendered from any unseen viewpoint.
Metric Based Few-Shot Graph Classification
In this work, we tackle the problem of few-shot graph classification, showing that equipping a simple distance metric learning baseline with a state-of-the-art graph embedder allows to obtain competitive results on the task.
Spectral Maps for Learning on Subgraphs
We demonstrate the benefits of incorporating spectral maps in graph learning pipelines, addressing scenarios where a node-to-node map is not well defined, or in the absence of exact isomorphism.
3D Human Pose Estimation Using Möbius Graph Convolutional Networks
In particular, this allows us to directly and explicitly encode the transformation between joints, resulting in a significantly more compact representation.
Ranked #71 on
3D Human Pose Estimation
on Human3.6M
Explanatory Learning: Beyond Empiricism in Neural Networks
We introduce Explanatory Learning (EL), a framework to let machines use existing knowledge buried in symbolic sequences -- e. g. explanations written in hieroglyphic -- by autonomously learning to interpret them.
Fish sounds: towards the evaluation of marine acoustic biodiversity through data-driven audio source separation
Moreover, one of the causes of biodiversity loss is sound pollution; in data obtained from regions with loud anthropic noise, it is hard to separate the artificial from the fish sound manually.
Graph Kernel Neural Networks
The convolution operator at the core of many modern neural architectures can effectively be seen as performing a dot product between an input matrix and a filter.
Smoothness and effective regularizations in learned embeddings for shape matching
In this study, we analyze, for the first time, properties that arise in data-driven learned embedding and their relation to the shape-matching task.
Unsupervised Source Separation via Bayesian Inference in the Latent Domain
State of the art audio source separation models rely on supervised data-driven approaches, which can be expensive in terms of labeling resources.
Ranked #1 on
Music Source Separation
on Slakh2100
Localized Shape Modelling with Global Coherence: An Inverse Spectral Approach
In this work, we address this problem by defining a data-driven model upon a family of linear operators (variants of the mesh Laplacian), whose spectra capture global and local geometric properties of the shape at hand.
Shape registration in the time of transformers
In this paper, we propose a transformer-based procedure for the efficient registration of non-rigid 3D point clouds.
Cluster-driven Graph Federated Learning over Multiple Domains
Clustering may reduce heterogeneity by identifying the domains, but it deprives each cluster model of the data and supervision of others.
Universal Spectral Adversarial Attacks for Deformable Shapes
Our attacks are universal, in that they transfer across different shapes, different representations (meshes and point clouds), and generalize to previously unseen data.
Learning Spectral Unions of Partial Deformable 3D Shapes
Spectral geometric methods have brought revolutionary changes to the field of geometry processing.
Learning disentangled representations via product manifold projection
We propose a novel approach to disentangle the generative factors of variation underlying a given set of observations.
Non-Rigid Puzzles
We assume to be given a reference shape and its multiple parts undergoing a non-rigid deformation.
High-Resolution Augmentation for Automatic Template-Based Matching of Human Models
This augmentation provides an effective workaround for the resolution limitations imposed by the adopted morphable model.
Multimodal Feature Fusion and Knowledge-Driven Learning via Experts Consult for Thyroid Nodule Classification
Computer-aided diagnosis (CAD) is becoming a prominent approach to assist clinicians spanning across multiple fields.
LIMP: Learning Latent Shape Representations with Metric Preservation Priors
In this paper, we advocate the adoption of metric preservation as a powerful prior for learning latent representations of deformable 3D shapes.
Instant recovery of shape from spectrum via latent space connections
We introduce the first learning-based method for recovering shapes from Laplacian spectra.
The Whole Is Greater Than the Sum of Its Nonrigid Parts
Here, we claim that observing part of an object which was previously acquired as a whole, one could deal with both partial matching and shape completion in a holistic manner.
ZoomOut: Spectral Upsampling for Efficient Shape Correspondence
Our main observation is that high quality maps can be obtained even if the input correspondences are noisy or are encoded by a small number of coefficients in a spectral basis.
Graphics
Self-supervised Learning of Dense Shape Correspondence
The resulting learning model is class-agnostic, and is able to leverage any type of deformable geometric data for the training phase.
Isospectralization, or how to hear shape, style, and correspondence
The question whether one can recover the shape of a geometric object from its Laplacian spectrum ('hear the shape of the drum') is a classical problem in spectral geometry with a broad range of implications and applications.
FARM: Functional Automatic Registration Method for 3D Human Bodies
We introduce a new method for non-rigid registration of 3D human shapes.
Efficient Deformable Shape Correspondence via Kernel Matching
We present a method to match three dimensional shapes under non-isometric deformations, topology changes and partiality.
Deep Functional Maps: Structured Prediction for Dense Shape Correspondence
We introduce a new framework for learning dense correspondence between deformable 3D shapes.
Product Manifold Filter: Non-Rigid Shape Correspondence via Kernel Density Estimation in the Product Space
Many algorithms for the computation of correspondences between deformable shapes rely on some variant of nearest neighbor matching in a descriptor space.
Geometric deep learning on graphs and manifolds using mixture model CNNs
Recently, there has been an increasing interest in geometric deep learning, attempting to generalize deep learning methods to non-Euclidean structured data such as graphs and manifolds, with a variety of applications from the domains of network analysis, computational social science, or computer graphics.
Ranked #4 on
Document Classification
on Cora
Bayesian Inference of Bijective Non-Rigid Shape Correspondence
Many algorithms for the computation of correspondences between deformable shapes rely on some variant of nearest neighbor matching in a descriptor space.
Learning shape correspondence with anisotropic convolutional neural networks
Establishing correspondence between shapes is a fundamental problem in geometry processing, arising in a wide variety of applications.
Efficient Globally Optimal 2D-to-3D Deformable Shape Matching
We propose the first algorithm for non-rigid 2D-to-3D shape matching, where the input is a 2D shape represented as a planar curve and a 3D shape represented as a surface; the output is a continuous curve on the surface.
Point-wise Map Recovery and Refinement from Functional Correspondence
Since their introduction in the shape analysis community, functional maps have met with considerable success due to their ability to compactly represent dense correspondences between deformable shapes, with applications ranging from shape matching and image segmentation, to exploration of large shape collections.
Partial Functional Correspondence
In this paper, we propose a method for computing partial functional correspondence between non-rigid shapes.
