Explanipedia

Advanced Layout Analysis Models for Docling Open

Nikolaos Livathinos, Christoph Auer, Ahmed Nassar, R. Teixeira De Lima, Maksym Lysak , et al. · 2025

This technical report documents the development of novel Layout Analysis models integrated into the Docling document-conversion pipeline. We trained several state-of-the-art object detectors based on the RT-DETR, RT-DETRv2 and DFINE archit…

Benchmarking Unsupervised Strategies for Anomaly Detection in Multivariate Time Series Open

Laura Boggia, R. Teixeira De Lima, Bogdan Malaescu · 2025

Anomaly detection in multivariate time series is an important problem across various fields such as healthcare, financial services, manufacturing or physics detector monitoring. Accurately identifying when unexpected errors or faults occur…

<span>ChemQuery</span>: A Natural Language Query‐Driven Service for Comprehensive Exploration of Chemistry Patent Literature Open

Shubham Gupta, R. Teixeira De Lima, Lokesh Mishra, Cèsar Berrospi, Panagiotis Vagenas , et al. · 2025

Patents are integral to our shared scientific knowledge, requiring companies and inventors to stay informed about them to conduct research, find licensing opportunities, and manage legal risks. However, the rising rate of filings has made …

SmolDocling: An ultra-compact vision-language model for end-to-end multi-modal document conversion Open

Ahmed Nassar, Andrés Marafioti, Matteo Omenetti, Maksym Lysak, Nikolaos Livathinos , et al. · 2025

We introduce SmolDocling, an ultra-compact vision-language model targeting end-to-end document conversion. Our model comprehensively processes entire pages by generating DocTags, a new universal markup format that captures all page element…

Granite Vision: a lightweight, open-source multimodal model for enterprise Intelligence Open

Granite Vision Team, Leonid Karlinsky, Assaf Arbelle, A Daniels, Ahmed Nassar , et al. · 2025

We introduce Granite Vision, a lightweight large language model with vision capabilities, specifically designed to excel in enterprise use cases, particularly in visual document understanding. Our model is trained on a comprehensive instru…

Docling: An Efficient Open-Source Toolkit for AI-driven Document Conversion Open

Nikolaos Livathinos, Christoph Auer, Maksym Lysak, Ahmed Nassar, Michele Dolfi , et al. · 2025

We introduce Docling, an easy-to-use, self-contained, MIT-licensed, open-source toolkit for document conversion, that can parse several types of popular document formats into a unified, richly structured representation. It is powered by st…

Know Your RAG: Dataset Taxonomy and Generation Strategies for Evaluating RAG Systems Open

R. Teixeira De Lima, Shubham Gupta, Cèsar Berrospi, Lokesh Mishra, Michele Dolfi , et al. · 2024

Retrieval Augmented Generation (RAG) systems are a widespread application of Large Language Models (LLMs) in the industry. While many tools exist empowering developers to build their own systems, measuring their performance locally, with d…

Docling Technical Report Open

Christoph Auer, Maksym Lysak, Ahmed Nassar, Michele Dolfi, Nikolaos Livathinos , et al. · 2024

This technical report introduces Docling, an easy to use, self-contained, MIT-licensed open-source package for PDF document conversion. It is powered by state-of-the-art specialized AI models for layout analysis (DocLayNet) and table struc…

INDUS: Effective and Efficient Language Models for Scientific Applications Open

Bishwaranjan Bhattacharjee, Aashka Trivedi, Masayasu Muraoka, Muthukumaran Ramasubramanian, Takuma Udagawa , et al. · 2024

Large language models (LLMs) trained on general domain corpora showed remarkable results on natural language processing (NLP) tasks. However, previous research demonstrated LLMs trained using domain-focused corpora perform better on specia…

Ultra-low latency recurrent neural network inference on FPGAs for physics applications with hls4ml Open

E. E. Khoda, Dylan Rankin, R. Teixeira De Lima, Philip Harris, Scott Hauck , et al. · 2023

Recurrent neural networks have been shown to be effective architectures for many tasks in high energy physics, and thus have been widely adopted. Their use in low-latency environments has, however, been limited as a result of the difficult…

Ultra-low latency recurrent neural network inference on FPGAs for physics applications with hls4ml Open

E. E. Khoda, Dylan Rankin, R. Teixeira De Lima, Philip Harris, Scott Hauck , et al. · 2022

Recurrent neural networks have been shown to be effective architectures for many tasks in high energy physics, and thus have been widely adopted. Their use in low-latency environments has, however, been limited as a result of the difficult…

Sequence-based Machine Learning Models in Jet Physics Open

R. Teixeira De Lima · 2021

Sequence-based modeling broadly refers to algorithms that act on data that is represented as an ordered set of input elements. In particular, Machine Learning algorithms with sequences as inputs have seen successfull applications to import…

Accessing HH at ATLAS and CMS with the HL-LHC Open

R. Teixeira De Lima · 2020

Directly measuring the Higgs self-coupling through Higgs pair production is an active area of interest for the LHC experiments, with a current limit around 10 times the Standard Model cross section per experiment. These results, exploting …

Deep Sets for Flavor Tagging on the ATLAS Experiment Open

N. M. Hartman, R. Teixeira De Lima, M. Kagan · 2020

Flavour Tagging is a major client for tracking in particle physics experiments at high energy colliders, where it is used to identify the experimental signatures of heavy flavor production. Among other features, charm and beauty hadron dec…

Higgs Boson Pair Production at Colliders: Status and Perspectives Open

B. Di Micco, M. Gouzevitch, Javier Mazzitelli, C. Vernieri, J. Alison , et al. · 2020

This document summarises the current theoretical and experimental status of the di-Higgs boson production searches, and of the direct and indirect constraints on the Higgs boson self-coupling, with the wish to serve as a useful guide for t…

Overview of Energy Reconstruction, and Electron and Photon Performances with the CMS ECAL in Run II Open

R. Teixeira De Lima · 2017

The electromagnetic calorimeter (ECAL) of the Compact Muon Solenoid (CMS) Experiment is crucial for achieving high resolution measurements of electrons and photons. Maintaining and possibly improving the excellent performance achieved in R…

Connectivity Maintenance of a Set of Agents through MST-based Algorithm Open

Cézanne Alves, Warley Gramacho, R. Teixeira De Lima · 2017

In this paper, it is proposed a solution to the problem of positioning a set of agents that play the role of pursing a set of moving targets, while the global connectivity among such agents is maintained throughout positioning a second set…

Optimizing Image Steganography using Particle Swarm Optimization Algorithm Open

R. Teixeira De Lima, Warley Gramacho, Ary Henrique · 2017

Image Steganography is the computing field of hiding information from a source into a target image in a way that it becomes almost imperceptible from one's eyes.Despite the high capacity of hiding information, the usual Least Significant B…

Beyond the standard model Higgs physics with photons with the CMS detector Open

R. Teixeira De Lima · 2017

The experimental discovery of the Higgs boson is one of the latest successes of the Standard Model of particle physics. Although all measurements have confirmed that this newly discovered particle is the Higgs boson predicted by the Standa…

R. Teixeira De Lima YOU? Author Swipe