Federico Bianchi

Federico Bianchi

Auto-generated profile, based on publicly available information

Federico Bianchi

Computer Science Dept., Stanford University, Stanford, USA

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 June 2023
Verified
Conference Paper
Open access
FBFederico Bianchi
Federico Bianchi
PKPratyusha Kalluri
Pratyusha Kalluri
EDEsin Durmus
Esin Durmus
FLFaisal Ladhak
Faisal Ladhak
MCMyra Cheng
Myra Cheng
ACAylin Caliskan
Aylin Caliskan
Machine learning models that convert user-written text descriptions into images are now widely available online and used by millions of users to generate millions of images a day. We investigate the potential for these models to amplify dangerous and complex stereotypes. We find a broad range of ordinary prompts produce stereotypes, including prompts simply mentioning traits, descriptors, occupations, or objects. For example, we find cases of prompting for basic traits or social roles resulting in images reinforcing whiteness as ideal, prompting for occupations resulting in amplification of racial and gender disparities, and prompting for objects resulting in reification of American norms. Stereotypes are present regardless of whether prompts explicitly mention identity and demographic language or avoid such language. Moreover, stereotypes persist despite mitigation strategies; neither user attempts to counter stereotypes by requesting images with specific counter-stereotypes nor institutional attempts to add system “guardrails” have prevented the perpetuation of stereotypes. Our analysis justifies concerns regarding the impacts of today’s models, presenting striking exemplars, and connecting these findings with deep insights into harms drawn from social scientific and humanist disciplines. This work contributes to the effort to shed light on the uniquely complex biases in language-vision models and demonstrates the ways that the mass deployment of text-to-image generation models results in mass dissemination of stereotypes and resulting harms.

17 July 2019 in Proceedings of the AAAI Conference on Artificial Intelligence
Verified
Journal Article
Closed access
PDF file
VCValerio Di Carlo
Valerio Di Carlo
FBFederico Bianchi
Federico Bianchi
MPMatteo Palmonari
Matteo Palmonari
Temporal word embeddings have been proposed to support the analysis of word meaning shifts during time and to study the evolution of languages. Different approaches have been proposed to generate vector representations of words that embed their meaning during a specific time interval. However, the training process used in these approaches is complex, may be inefficient or it may require large text corpora. As a consequence, these approaches may be difficult to apply in resource-scarce domains or by scientists with limited in-depth knowledge of embedding models. In this paper, we propose a new heuristic to train temporal word embeddings based on the Word2vec model. The heuristic consists in using atemporal vectors as a reference, i.e., as a compass, when training the representations specific to a given time interval. The use of the compass simplifies the training process and makes it more efficient. Experiments conducted using state-of-the-art datasets and methodologies suggest that our approach outperforms or equals comparable approaches while being more robust in terms of the required corpus size.

6 February 2021 in Applied Intelligence
Verified
Journal Article
Closed access
PDF file
MEMonireh Ebrahimi
Monireh Ebrahimi
AEAaron Eberhart
Aaron Eberhart
FBFederico Bianchi
Federico Bianchi
PHPascal Hitzler
Pascal Hitzler

1 November 2020
Verified
Book Chapter
Closed access
PDF file
VCVincenzo Cutrona
Vincenzo Cutrona
FBFederico Bianchi
Federico Bianchi
EJErnesto Jiménez-Ruiz
Ernesto Jiménez-Ruiz
MPMatteo Palmonari
Matteo Palmonari

1 April 2023 in bioRxiv
Verified
Preprint
Open access
PDF file
ZHZhi Huang
Zhi Huang
FBFederico Bianchi
Federico Bianchi
MYMert Yuksekgonul
Mert Yuksekgonul
TMThomas Montine
Thomas Montine
JZJames Y Zou
James Y Zou
The lack of annotated publicly available medical images is a major barrier for innovations. At the same time, many de-identified images and much knowledge are shared by clinicians on public forums such as medical Twitter. Here we harness these crowd platforms to curate OpenPath, a large dataset of 208,414 pathology images paired with natural language descriptions. This is the largest public dataset for pathology images annotated with natural text. We demonstrate the value of this resource by developing PLIP, a multimodal AI with both image and text understanding, which is trained on OpenPath. PLIP achieves state-of-the-art zero-shot and transfer learning performances for classifying new pathology images across diverse tasks. Moreover, PLIP enables users to retrieve similar cases by either image or natural language search, greatly facilitating knowledge sharing. Our approach demonstrates that publicly shared medical information is a tremendous resource that can be harnessed to advance biomedical AI.

10 November 2021 in Physical Review A
Verified
Journal Article
Closed access
PDF file
APAdriano M. Palmieri
Adriano M. Palmieri
FBFederico Bianchi
Federico Bianchi
MPMatteo Paris
Matteo Paris
CBClaudia Benedetti
Claudia Benedetti
We address the use of neural networks (NNs) in classifying the environmental parameters of single-qubit dephasing channels. In particular, we investigate the performance of linear perceptrons and of two nonlinear NN architectures. At variance with time-series-based approaches, our goal is to learn a discretized probability distribution over the parameters using tomographic data at just two random instants of time. We consider dephasing channels originating either from classical 1/fα noise or from the interaction with a bath of quantum oscillators. The parameters to be classified are the color α of the classical noise or the Ohmicity parameter s of the quantum environment. In both cases, we find that NNs are able to exactly classify parameters into 16 classes using noiseless data (a linear NN is enough for the color, whereas a single-layer NN is needed for the Ohmicity). In the presence of noisy data (e.g., coming from noisy tomographic measurements), the network is able to classify the color of the 1/fα noise into 16 classes with about 70% accuracy, whereas classification of Ohmicity turns out to be challenging. We also consider a more coarse-grained task and train the network to discriminate between two macroclasses corresponding to α≶1 and s≶1, obtaining up to 96% and 79% accuracy using single-layer NNs.