home.social

#slurs — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #slurs, aggregated by home.social.

fetched live
  1. Odds are, the guy with the headphones wasn’t even #gay. So I had to infer the shouting man would use that slur to describe anyone who diverges from his narrow view of masculinity — But then what exactly set him off?
    medium.com/prismnpen/you-screa

    #LGBTQ #Slurs #Words #Language

  2. A-K-M-E – Slurs come segnali linguistici

    Piano Terra, viernes, 29 de mayo, 21:00 CEST

    A-K-M-E – Slurs come segnali linguistici

    t.me/akme_lab

    29.05 – dalle ore 21:00

    Tra NLP, modelli transformer e dinamiche algoritmiche

    Nel contesto dei social network, gli slurs, ovvero gli insulti e i concetti discriminatori operano come feature linguistiche ad alta entropia, capaci di perturbare in modo significativo i modelli NLP basati su architetture transformer.

    Analizziamo assieme cercando di comprendere il fenomeno da una prospettiva linguistica computazionale, osservando come i token tossici influenzino embeddings, attention patterns, loss surfaces e sistemi di ranking.

    L’obiettivo è mostrare come la tossicità linguistica sia un effetto emergente dell’interazione tra strutture linguistiche, architetture transformer, dataset imperfetti e metriche di ottimizzazione.

    – failure modes dei modelli di rilevazione tossicità: bias nei dataset, overfitting su pattern superficiali, misclassificazione in contesti ironici o citazionali

    – analisi delle attention heads: heads specializzate nella rilevazione di segnali tossici, saturazione dell’attenzione e perdita di contesto pragmatico

    – limitazioni delle context windows: difficoltà nel disambiguare uso descrittivo vs offensivo, soprattutto in sequenze brevi o rumorose

    – propagation bias nei sistemi di raccomandazione: amplificazione dei token ad alta reattività, feedback loop tra engagement e ranking, distorsione delle distribuzioni linguistiche

    – impatto sulle loss functions: penalizzazioni non uniformi, gradient explosion in presenza di token tossici, difficoltà nella calibratura dei modelli di moderazione

    29 maggio a partire dalle 21:00

    akme.vado.li

    hacker.convoca.la/event/a-k-m-

  3. It's cute how fast #baby #communists in the #West went from freeing #Palestine and making racial #slurs about #Ukrainians to supporting the #Chinese regime and denying the #TiananmenSquare #massacre.

    And all because of freaking #TikTok.

  4. we dont fucking needs #slurs for anything if you think we need slurs for something you need to #fuckoff.

  5. oh my gawd you’re an #abled white lady and you think calling me #slurs to my face is not only your right but a #feminist act in defense of women???? that’s so original #nobody’s EVER done that before….

  6. The Register: Sora makes slurfect deepfakes of celebs spewing racial epithets . “Guardrails? What guardrails? Naughty netizens found a way to trick the Sora 2 video generator into producing deepfakes of public figures, including OpenAI CEO Sam Altman and billionaire Mark Cuban, that make it sound as though they’re spewing racial slurs. The trick works despite Sora’s built-in filters meant to […]

    https://rbfirehose.com/2025/10/25/the-register-sora-makes-slurfect-deepfakes-of-celebs-spewing-racial-epithets/

  7. Comedy, especially in satire, relies on a delicate balance between intent and interpretation. The intent behind a joke matters deeply, more than most people give credit for. In the case of MeatCanyon, his work thrives on exaggeration and absurdity, not cruelty.
    medium.com/prismnpen/im-queer-

    #LGBTQ #Queer #Slurs #FreeSpeech #Reclamation

  8. Dan Pfeiffer: "I understand that the Hitler references in this Young Republicans group chat may not seem so bad to the guy (Vice President JD Vance) who is now undyingly loyal to a man he once described as "America's Hitler."
    #racism #slurs #values #politics

  9. #OpenAI Designed GPT-5 to Be Safer. It Still Outputs #Gay #Slurs

    The new version of #ChatGPT explains why it won’t generate rule-breaking outputs. WIRED’s initial analysis found that some guardrails were easy to circumvent.
    #ai #artificialintelligence #gpt5

    wired.com/story/openai-gpt5-sa

  10. In October 2004, #FrankJudeJr. was savagely beaten by a mob of off-duty #Milwaukee #police officers who falsely accused him of stealing a police badge. The #officers screamed #racial #slurs at him during the #attack. When on-duty officers arrived, they either did nothing or joined the attack.

  11. No one acknowledged or cared that the man was being called the F-slur. No one could forgive him for the awful thing he’d done… and fair enough... but it was as if the slur was what he “deserved”.
    buff.ly/11O8HTa

    #LGBTQ #Assault #Slurs #Bigotry

  12. "The term ["moron"] is attributed to psychologist and eugenicist Henry H. Goddard, who used it to describe “feeble-minded” individuals. It is closely tied to the United States’s involvement in eugenics [...]"

    Yeah, I know many on here already avoid using ableist language like this, but definitely good to know the history of this particular term.

    teenvogue.com/story/the-sinist

    via mas.to/@meganL/114108269406867

    #ableism #slurs #language #WordsMatter

  13. PsyPost: Google searches for racial slurs are higher in areas where people are worried about disease. “A new study explores how pathogen stress—the prevalence of infectious diseases—may fuel racial bias in the United States, particularly anti-Black prejudice. By analyzing Google search data and measures of implicit and explicit bias, the researchers found that while pathogen stress […]

    https://rbfirehose.com/2025/01/13/psypost-google-searches-for-racial-slurs-are-higher-in-areas-where-people-are-worried-about-disease/