FBFederico Bianchi

Federico Bianchi

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Federico Bianchi has a total of 86 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Author Correction: A challenge for rounded evaluation of recommender systems

    by Jacopo Tagliabue, Federico Bianchi, Tobias Schnabel, Giuseppe Attanasio, Ciro Greco, Gabriel de Souza Moreira, Patrick John Chia

  • Open Access

    “Keep Your Heads Held High Boys!”: Examining the Relationship between the Proud Boys’ Online Discourse and Offline Activities

    by Catie Snow Bailard, Rebekah Tromble, Wei Zhong, Federico Bianchi, Pedram Hosseini, David Broniatowski

    This study examines the relationship between online communication by the Proud Boys and their offline activities. We use a supervised machine learning model to analyze a novel dataset of Proud Boys Telegram messages, merged with US Crisis Monitor data of violent and nonviolent events in which group members participated over a 31-month period. Our analysis finds that intensifying expressions of grievances online predict participation in offline violence, whereas motivational appeals to group pride, morale, or solidarity share a reciprocal relationship with participation in offline events. This suggests a potential online messaging–offline action cycle, in which (a) nonviolent offline protests predict an increasing proportion of motivational messaging and (b) increases in the frequency and proportion of motivational appeals online, in turn, predict subsequent violent offline activities. Our findings offer useful theoretical insights for understanding the relationship between online speech and offline behavior.


  • Open Access

    Meaning Modulations and Stability in Large Language Models: An Analysis of BERT Embeddings for Psycholinguistic Research

    by Giovanni Cassani, Federico Bianchi, Giuseppe Attanasio, Marco Marelli, Fritz Guenther

    Computational models of semantic representations have long assumed and produced a single static representation for each word type, ignoring the influence of linguistic context on semantic representations. Recent Large Language Models (LLMs) introduced in Natural Language Processing, however, learn token-level contextualised representations, holding promise to study how semantic representations change in different contexts. In this study we probe type- and token-level representations learned using a prominent example of such models, Bidirectional Encoder Representations from Transformers (BERT), for their ability to i) explain semantic effects found for isolated words (semantic relatedness and similarity ratings, lexical decision, and semantic priming), but critically also to ii) exhibit systematic interactions between lexical semantics and context, and iii) explain meaning modulations in context. Across a wide range of empirical studies on each of these topics, we show that BERT representations satisfy two desiderata for psychologically valid semantic representations: i) they have a stable semantic core which allows people to interpret words in isolation and prevents words to be used arbitrarily and ii) they interact with sentence context in systematic ways, with representations shifting as a function of their semantic core and the context. This demonstrates that a single, comprehensive model which simultaneously learns abstract, type-level prototype representations as well as mechanisms of how these interact with context can explain both isolated word effects and context-dependent variations. Notably, these variations are not limited to discrete word senses, eschewing a strict dichotomy between exemplar and prototype models and re-framing traditional notions of polysemy in semantic memory.


  • Open Access

    EvalRS 2023: Well-Rounded Recommender Systems for Real-World Deployments

    by Federico Bianchi, Patrick John Chia, Jacopo Tagliabue, Ciro Greco, Gabriel S P Moreira, Davide Eynard, Fahd Husain, Claudio Pomo

    EvalRS aims to bring together practitioners from industry and academia to foster a debate on rounded evaluation of recommender systems, with a focus on real-world impact across a multitude of deployment scenarios. Recommender systems are often evaluated only through accuracy metrics, which fall short of fully characterizing their generalization capabilities and miss important aspects, such as fairness, bias, usefulness, informativeness. This workshop builds on the success of last year's workshop at CIKM, but with a broader scope and an interactive format.


  • Open Access

    EvalRS 2023. Well-Rounded Recommender Systems For Real-World Deployments

    by Federico Bianchi, Patrick John Chia, Ciro Greco, Claudio Pomo, Gabriel Moreira, Davide Eynard, Fahd Husain, Jacopo Tagliabue

    EvalRS aims to bring together practitioners from industry and academia to foster a debate on rounded evaluation of recommender systems, with a focus on real-world impact across a multitude of deployment scenarios. Recommender systems are often evaluated only through accuracy metrics, which fall short of fully characterizing their generalization capabilities and miss important aspects, such as fairness, bias, usefulness, informativeness. This workshop builds on the success of last year's workshop at CIKM, but with a broader scope and an interactive format.

  • Top Cited Papers
    Open Access

    Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scale

    by Federico Bianchi, Pratyusha Kalluri, Esin Durmus, Faisal Ladhak, Myra Cheng, Debora Nozza, Tatsunori Hashimoto, Dan Jurafsky, James Zou, 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.


  • Open Access

    Is It Worth the (Environmental) Cost? Limited Evidence for Temporal Adaptation via Continuous Training

    by Giuseppe Attanasio, Debora Nozza, Federico Bianchi, Dirk Hovy

    Language is constantly changing and evolving, leaving language models to become quickly outdated. Consequently, we should continuously update our models with new data to expose them to new events and facts. However, that requires additional computing, which means new carbon emissions. Do any measurable benefits justify this cost? This paper looks for empirical evidence to support continuous training. We reproduce existing benchmarks and extend them to include additional time periods, models, and tasks. Our results show that the downstream task performance of temporally adapted English models for social media data do not improve over time. Pretrained models without temporal adaptation are actually significantly more effective and efficient. However, we also note a lack of suitable temporal benchmarks. Our findings invite a critical reflection on when and how to temporally adapt language models, accounting for sustainability.

  • Open Access

    Leveraging medical Twitter to build a visual–language foundation model for pathology AI

    by Zhi Huang, Federico Bianchi, Mert Yuksekgonul, Thomas Montine, James 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.


  • Open Access

    Beyond Digital "Echo Chambers": The Role of Viewpoint Diversity in Political Discussion

    by Rishav Hada, Amir Ebrahimi Fard, Sarah Shugars, Federico Bianchi, Patricia Rossini, Dirk Hovy, Rebekah Tromble, Nava Tintarev

    Increasingly taking place in online spaces, modern political conversations are typically perceived to be unproductively affirming---siloed in so called "echo chambers" of exclusively like-minded discussants. Yet, to date we lack sufficient means to measure viewpoint diversity in conversations. To this end, in this paper, we operationalize two viewpoint metrics proposed for recommender systems and adapt them to the context of social media conversations. This is the first study to apply these two metrics (Representation and Fragmentation) to real world data and to consider the implications for online conversations specifically. We apply these measures to two topics---daylight savings time (DST), which serves as a control, and the more politically polarized topic of immigration. We find that the diversity scores for both Fragmentation and Representation are lower for immigration than for DST. Further, we find that while pro-immigrant views receive consistent pushback on the platform, anti-immigrant views largely operate within echo chambers. We observe less severe yet similar patterns for DST. Taken together, Representation and Fragmentation paint a meaningful and important new picture of viewpoint diversity.


  • Gender and Age Bias in Commercial Machine Translation

    by Federico Bianchi, Tommaso Fornaciari, Dirk Hovy, Debora Nozza