Poincaré disk model
PulseAugur coverage of Poincaré disk model — every cluster mentioning Poincaré disk model across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
New adaptive curve subdivision method learns local angles for improved geometry generation
Researchers have developed a new adaptive interpolatory curve subdivision method that learns local angles to improve geometric object generation. This approach ensures the refined curve passes through control points whi…
-
New research maps olfactory data in hyperbolic space
Researchers have explored the organization of olfactory descriptor data using a two-dimensional hyperbolic embedding within the Poincaré disk model. By applying hyperbolic metric multidimensional scaling to datasets of …
-
HypLTSF framework uses hyperbolic geometry for advanced time series forecasting
Researchers have developed HypLTSF, a novel framework that models multi-scale hierarchies in time series forecasting using hyperbolic geometry. By embedding scale-wise representations into the Poincaré ball, HypLTSF nat…
-
ISO-RAG framework enhances retrieval for complex question answering
Researchers have introduced ISO-RAG, a novel framework for retrieval-augmented generation (RAG) that addresses limitations in multi-hop question answering. By leveraging a hyperbolic Poincaré disk model, ISO-RAG prunes …
-
New framework enhances attention retrieval in hyperbolic spaces
Researchers have developed HCC+, a theoretical framework for improving attention retrieval in hyperbolic spaces. This new method addresses the lack of deterministic guarantees in existing techniques regarding attention-…
-
Hyperbolic geometry boosts tree-structured prototype networks
Researchers have explored the impact of latent manifold geometry on hierarchical classification models, comparing Euclidean and hyperbolic spaces. Their findings indicate that hyperbolic prototypes significantly preserv…
-
Hyperbolic geometry boosts latent space topology in classification models
Researchers explored the impact of latent manifold choice on hierarchical classification models, comparing Euclidean space with hyperbolic space (Poincaré ball). Their findings indicate that hyperbolic prototypes signif…
-
HyPASE framework uses hyperbolic geometry for efficient LALM fine-tuning
Researchers have developed HyPASE, a novel framework that utilizes hyperbolic geometry for parameter-efficient fine-tuning of Large Audio-Language Models (LALMs) for Speech Emotion Recognition (SER). Unlike traditional …
-
New geometric construction method for neural networks detailed in arXiv paper
Researchers have developed a novel method for constructing neural networks by embedding them within statistical manifolds, specifically utilizing the lognormal distribution. This approach leverages a Hamiltonian system …
-
New hyperbolic framework enhances multimodal emotion and sentiment analysis
Researchers have introduced Emotion Collider (EC-Net), a novel hyperbolic hypergraph framework designed for multimodal emotion and sentiment analysis. This system utilizes Poincaré-ball embeddings to model modality hier…
-
New 1-Lipschitz Neural Networks Developed for Hadamard Manifolds
Researchers have developed a new class of 1-Lipschitz neural networks designed to operate on Hadamard manifolds, moving beyond the limitations of Euclidean spaces. These networks utilize Busemann functions and gradient …
-
GPT-2 vocabulary visualized as interactive hyperbolic tree
A follow-up project visualizes the vocabulary of the GPT-2 model as a hyperbolic tree, containing 32,070 tokens within a navigable Poincaré disk. This interactive visualization, runnable on mobile devices, uses token em…
-
New AI Framework Fuses Infrared and Visible Images Using Hyperbolic Geometry
Researchers have developed a novel framework for fusing infrared and visible images by leveraging hyperbolic manifold learning. This approach uses text prompts, extracted by BLIP, as anchors in hyperbolic space to align…