arXiv subjects

Radames Cruz Moreno

Publications and source records attributed to Radames Cruz Moreno.

4 recordsLinked to original sources

arXiv2025

Sandi: A System for Accountability

We present a system, Sandi, for creating trust through accountability. Concretely, we focus on online communication scenarios, where the communicating parties do not know each other, yet would benefit from a degree of initial trust. Sandi can be seen as a reputation system that measures bad behavior, with strong integrity protections and resistance to manipulation. Unlike most reputation systems, Sandi is entirely based on ``downvotes'' and therefore requires strong privacy guarantees to prevent retaliation. It utilizes a ticket-based reporting mechanism to limit who can report. We also prove that Sandi incentivizes good behavior in a well-defined sense. Sandi is by design unidirectional, so that message senders have Sandi scores and receivers can report them for inappropriate communication, but it is designed to benefit both senders and receivers. Senders benefit, as receivers are more likely to react to communication with the added trust signal. Receivers benefit from seeing senders' scores, allowing them to make more informed decisions about which senders to trust. Receivers do not need registered accounts and neither senders nor receivers need long-term keys. Sandi guarantees score integrity, communication privacy, reporter privacy to protect reporting receivers, and sender unlinkability. Sandi can be implemented on top of any communication system that allows for small binary data transfer.

cs.CR↗

arXiv2023

Sandi: A System for Accountability and Applications in Direct Communication (Extended Abstract)

Reputation systems guide our decision making both in life and work: which restaurant to eat at, which vendor to buy from, which software dependencies to use, and who or what to trust. These systems are often based on old ideas and are failing in the face of modern threats. Fraudsters have found ways to manipulate them, undermining their integrity and utility. Generative AI adds to the problem by enabling the creation of real-looking fake narratives at scale, creating a false sense of consensus. Meanwhile, the need for reliable reputation concepts is more important than ever, as wrong decisions lead to increasingly severe outcomes: wasted time, poor service, and a feeling of injustice at best, fraud, identity theft, and ransomware at worst. In this extended abstract we introduce Sandi, a new kind of reputation system with a single well-defined purpose: to create trust through accountability in one-to-one transactions. Examples of such transactions include sending an email or making a purchase online. Sandi has strong security and privacy properties that make it suitable for use also in sensitive contexts. Furthermore, Sandi can guarantee reputation integrity and transparency for its registered users. As a primary application, we envision how Sandi could counter fraud and abuse in direct communication. Concretely, message senders request a cryptographic tag from Sandi that they send along with their message. If the receiver finds the message inappropriate, they can report the sender using this tag. Notably, only senders need registered accounts and do not need to manage long-term keys. The design of Sandi ensures compatibility with any communication system that allows for small binary data transmission.

cs.CR↗

arXiv2023

Improving the Reusability of Pre-trained Language Models in Real-world Applications

The reusability of state-of-the-art Pre-trained Language Models (PLMs) is often limited by their generalization problem, where their performance drastically decreases when evaluated on examples that differ from the training dataset, known as Out-of-Distribution (OOD)/unseen examples. This limitation arises from PLMs' reliance on spurious correlations, which work well for frequent example types but not for general examples. To address this issue, we propose a training approach called Mask-tuning, which integrates Masked Language Modeling (MLM) training objectives into the fine-tuning process to enhance PLMs' generalization. Comprehensive experiments demonstrate that Mask-tuning surpasses current state-of-the-art techniques and enhances PLMs' generalization on OOD datasets while improving their performance on in-distribution datasets. The findings suggest that Mask-tuning improves the reusability of PLMs on unseen data, making them more practical and effective for real-world applications.

cs.CL↗

arXiv2023

Gender-tuning: Empowering Fine-tuning for Debiasing Pre-trained Language Models

Recent studies have revealed that the widely-used Pre-trained Language Models (PLMs) propagate societal biases from the large unmoderated pre-training corpora. Existing solutions require debiasing training processes and datasets for debiasing, which are resource-intensive and costly. Furthermore, these methods hurt the PLMs' performance on downstream tasks. In this study, we propose Gender-tuning, which debiases the PLMs through fine-tuning on downstream tasks' datasets. For this aim, Gender-tuning integrates Masked Language Modeling (MLM) training objectives into fine-tuning's training process. Comprehensive experiments show that Gender-tuning outperforms the state-of-the-art baselines in terms of average gender bias scores in PLMs while improving PLMs' performance on downstream tasks solely using the downstream tasks' dataset. Also, Gender-tuning is a deployable debiasing tool for any PLM that works with original fine-tuning.

cs.CL↗