Matias Travizano

Bruno Bianchi, Diego Tiscornia, Matias Travizano et al.

Since the advent of Large Language Models (LLMs), a significant area of research has focused on their intrinsic biases, particularly in political discourse. This study investigates a different but related concept, "political plasticity", which is defined as the capacity of models to adapt their responses based on the user supplied context. To analyze this, a testing framework was developed using an expanded corpus of 200 politically-oriented questions across economic and personal freedom axes, based on a prior framework by Lester (1996). The study explored several methods to induce political bias, including simplified and topic-based system prompts, as well as user prompts with few-shot examples. The results show that while system prompts were largely ineffective, user prompts successfully elicited significant ideological shifts, particularly along the Economic Freedom axis in larger and newer models. Through a validation experiment, we examined whether models answer questionnaires by recognizing the underlying question format. Inverting the sense of the questions revealed unexpected, counter-intuitive shifts in most models, suggesting potential data leakage. Finally, we also analyzed how model plasticity varies when the experiment is conducted in different languages. The results reveal subtle yet notable shifts across each of the analyzed languages. Overall, our results indicate that small and older LLMs exhibit limited or unstable political plasticity, whereas newer frontier models display reliable, expected adaptability.

Ariel Futoransky, Carlos Sarraute, Daniel Fernandez et al.

We construct a privacy-preserving, distributed and decentralized marketplace where parties can exchange data for tokens. In this market, buyers and sellers make transactions in a blockchain and interact with a third party, called notary, who has the ability to vouch for the authenticity and integrity of the data. We introduce a protocol for the data-token exchange where neither party gains more information than what it is paying for, and the exchange is fair: either both parties gets the other's item or neither does. No third party involvement is required after setup, and no dispute resolution is needed.

Ariel Futoransky, Carlos Sarraute, Ariel Waissbein et al.

We present a cryptographic primitive called WibsonTree designed to preserve users' privacy by allowing them to demonstrate predicates on their personal attributes, without revealing the values of those attributes. We suppose that there are three types of agents --buyers, sellers and notaries-- who interact in a decentralized privacy-preserving data marketplace (dPDM) such as the Wibson marketplace. We introduce the WibsonTree protocol as an efficient cryptographic primitive that enables the exchange of private information while preserving the seller's privacy. Using our primitive, a data seller can efficiently prove that he/she belongs to the target audience of a buyer's data request, without revealing any additional information.

Daniel Fernandez, Ariel Futoransky, Gustavo Ajzenman et al.

Wibson is a blockchain-based, decentralized data marketplace that provides individuals a way to securely and anonymously sell information in a trusted environment. The combination of the Wibson token and blockchain-enabled smart contracts hopes to allow Data Sellers and Data Buyers to transact with each other directly while providing individuals the ability to maintain anonymity as desired. The Wibson marketplace will provide infrastructure and financial incentives for individuals to securely sell personal information without sacrificing personal privacy. Data Buyers receive information from willing and actively participating individuals with the benefit of knowing that the personal information should be accurate and current. We present here two different components working together to achieve an efficient decentralized marketplace. The first is a smart contract called Data Exchange, which stores references to Data Orders that different Buyers open in order to show to the market that they are interested in buying certain types of data, and provides secure mechanisms to perform the transactions. The second is used to process payments from Buyers to Sellers and intermediaries, and is called Batch Payments.

Ariel Futoransky, Carlos Sarraute, Ariel Waissbein et al.

We are tackling the problem of trading real-world private information using only cryptographic protocols and a public blockchain to guarantee honest transactions. In this project, we consider three types of agents --buyers, sellers and notaries-- interacting in a decentralized privacy-preserving data marketplace (dPDM) such as the Wibson data marketplace. This framework offers infrastructure and financial incentives for individuals to securely sell personal information while preserving personal privacy. Here we provide an efficient cryptographic primitive for the secure exchange of data in a dPDM, which occurs as an atomic operation wherein the data buyer gets access to the data and the data seller gets paid simultaneously.

Matias Travizano, Carlos Sarraute, Gustavo Ajzenman et al.

Our aim is for Wibson to be a blockchain-based, decentralized data marketplace that provides individuals a way to securely and anonymously sell information in a trusted environment. The combination of the Wibson token and blockchain-enabled smart contracts hopes to allow Data Sellers and Data Buyers to transact with each other directly while providing individuals the ability to maintain anonymity as desired. Wibson intends that its data marketplace will provide infrastructure and financial incentives for individuals to securely sell personal information without sacrificing personal privacy. Data Buyers receive information from willing and actively participating individuals with the benefit of knowing that the personal information should be accurate and current.

Martin Fixman, Ariel Berenstein, Jorge Brea et al.

The explosion of mobile phone communications in the last years occurs at a moment where data processing power increases exponentially. Thanks to those two changes in a global scale, the road has been opened to use mobile phone communications to generate inferences and characterizations of mobile phone users. In this work, we use the communication network, enriched by a set of users' attributes, to gain a better understanding of the demographic features of a population. Namely, we use call detail records and banking information to infer the income of each person in the graph.

Juan de Monasterio, Alejo Salles, Carolina Lang et al.

Chagas disease is a neglected disease, and information about its geographical spread is very scarse. We analyze here mobility and calling patterns in order to identify potential risk zones for the disease, by using public health information and mobile phone records. Geolocalized call records are rich in social and mobility information, which can be used to infer whether an individual has lived in an endemic area. We present two case studies in Latin American countries. Our objective is to generate risk maps which can be used by public health campaign managers to prioritize detection campaigns and target specific areas. Finally, we analyze the value of mobile phone data to infer long-term migrations, which play a crucial role in the geographical spread of Chagas disease.