Papel is a social network for researchers. Discover, read, and engage with academic papers through AI-powered summaries, quizzes, and a community of researchers — all in one beautiful app. Built with on-device AI for complete privacy.

A space for my thoughts, research notes, and articles about technology, art, and everything in between.

This project studies the "Seats in Room" problem as a declarative optimization task, where students must be partitioned into groups and assigned to seats under social, spatial, and relaxable constraints. I modeled the problem with Answer Set Programming using Clingo and Constraint Programming using MiniZinc/Gecode, precomputing Euclidean seat distances to avoid non-linear solver overhead during search. The report compares optimization hierarchies, search heuristics, timeout behavior, and solution quality, showing how conflict-driven ASP solving scales better on tightly constrained logical instances while lightweight CP heuristics remain useful as a reproducible baseline.

This project investigates few-shot semantic segmentation for industrial visual inspection, focusing on semiconductor wafer defect images where labeled masks are scarce and domain expertise is costly. The work combines a technical literature review of conditional networks, prototype-based methods, latent-space optimization, contrastive learning, and vision foundation models with the implementation of a generalist Transformer-based segmentation pipeline. I experimented with SegGPT through Hugging Face, using prompt images and masks to perform one-shot defect segmentation, and compared the approach with SAM2 prompting to analyze segmentation quality, prompt sensitivity, inference cost, and deployment trade-offs for AI-assisted quality control.

This research provides a small but complete (at least, I tried) overview and analysis of the rapidly evolving field of Visual Language Models (VLMs). It’s hard to miss the exponential growth in Vision-Language Models (VLMs) since Radford et al. dropped CLIP back in 2021 (you can literally see it in Figure 1 of the paper). Suddenly, models combining vision and language were everywhere: ALIGN, CoCa, FLAVA, LiT, SLIP, generative models joining the fray... While incredibly exciting, it quickly became overwhelming. As someone trying to keep up and understand the real progress, I found myself struggling. It felt like a flood of new acronyms and techniques, each claiming state-of-the-art, but often making it difficult to grasp the fundamental differences and trade-offs beyond specific benchmark results.

Nuclear energy emerged out of atomic science advancements during WWII and enjoyed several decades of significant expansion to address industrializing nations’ energy demands. Over many years, nuclear energy saw significant growth worldwide; countries invested heavily in nuclear infrastructure, using its power to drive advancements in cities, medicine, and industry. Nuclear energy appeared limitless in its potential, sparking imaginations everywhere. The data used to develop this project come from Ember’s Yearly Electricity Data; Ember’s European Electricity Review; Energy Institute Statistical Review of World Energy and ElectricityMap.

During 10 months spent in Spain for the Erasmus experience I worked on a research project with the title 'Predicting the reactivity pulse values of a TRIGA nuclear reactor, a neural networks approach'. This research aims to develop and apply a neural network in the field of nuclear engineering; using Deep Learning algorithms it is possible to increase the safety of reactors and facilitate the work done by the technicians.

Data science is an exciting discipline that allows you to turn raw data into understanding, insight, and knowledge. **Hadley Wickham**
In june 2021 I shared my data science project concerning the psychological effects of the covid-19 pandemic throughout a specific lockdown period, entitled 'covidstress-project'. Evaluated in the examination with the highest judgment value, 30 cum lauda.

Between May and June 2022 I participated in a team competition proposed by the machine learning course teacher in which the goal was to use machine learning techniques and models to seek a solution to the problem of Income Prediction. My team won the competition, achieving an accuracy value of 0.87 using supervised learning techniques.

Deep Learning techniques also turn out to be essential to help medical doctors in diagnosing certain diseases. For example, breast cancer can be diagnosed through classification algorithms with good accuracy. I contributed to the research through an executable notebook from the popular Kaggle platform.

In Erasmus I took a university course in autonomous robotic in which as a final exam with my team we presented a project entitled "Development of a Robot for Hazardous Environment Tasks"; within the course we completed several tasks including the development of Kalman and particle filters using python as programming language and finally the development of proportional-integral-derivative controllers to manage the robot's movements in space.

My connection with technology began in 2013 when I started to understand how an android operating system works, therefore I started to create my own custom roms using as base the source codes of some popular aftermarket firmwares. In that time I met many people who contributed to the growth of my skills, @gokulbalram was my first guru and he deserves a special thanks. Like everything on this earth, this chapter of my life also came to an end, in 2017, after completing my latest job: an AOSP Porting for the ASUS Zenfone 2, powered by an Intel CPU. With the beginning of my university career, I deepened data science, generative and crypto art, blockchain and ethics related to technology. I like to work in dynamic, social environments where my creativity is encouraged. I believe that great goals can be achieved by reaching out to others, sharing your vision and valuing their input. The fundamental question I ask myself is: can an artificial intelligence in a near future think and act like a human? Make rational choices? Feels emotions?
being updated ...