#multimodality — Public Fediverse posts

🎥 Academic publishing beyond the written article

The new peer-reviewed Audio/Visual Section of Mediální studia / Media Studies is inviting submissions for its inaugural 2027 issue.

Film, video, audiovisual and sound essays are welcome, particularly projects bringing academic research into dialogue with artistic and multimodal practices.

📅 Deadline: 15 September 2026

🔗 https://lnkd.in/e97ps3ZR

#CFP #MediaStudies #Multimodality #AcademicPublishing

🎥 Academic publishing beyond the written article

The new peer-reviewed Audio/Visual Section of Mediální studia / Media Studies is inviting submissions for its inaugural 2027 issue.

Film, video, audiovisual and sound essays are welcome, particularly projects bringing academic research into dialogue with artistic and multimodal practices.

📅 Deadline: 15 September 2026

🔗 https://lnkd.in/e97ps3ZR

#CFP #MediaStudies #Multimodality #AcademicPublishing

Kandinsky WM 1.0: генеративные модели для Physical AI

В конце прошлого года мы представили семейство моделей генерации изображений и видео Kandinsky 5.0. Флагманская модель Kandinsky 5.0 Video Pro заняла и до сих пор удерживает первое место среди open-source-моделей в категории text-to-video на arena.ai . Сегодня мы выкладываем в открытый доступ Kandinsky WM 1.0, набор из трех специализированных моделей генерации видео для доменов Physical AI: автомобильные сценарии, робототехника и сцены со сложной физикой.

https://habr.com/ru/companies/sberbank/articles/1054966/

#physical_ai #genai #генерация_видео #kandinsky #multimodality #generative_models #texttovideo #world_models #robotics #selfdrivingcar

It was a complicated day for logistics. Note the motorcycle helmet in the bike bag in the back of the taxi with my folding bike.

Honestly, I don't know how some car-free and low-car people get by without folding bikes.

#Africa #Madagascar #Antananarivo #Malagasy #Brompton #Multimodality #Carfree #LowCar #Bike #Cycling #CyclingLife #CommuteByBike #NotJustBikes #Urbanism #Photography #OriginalPhoto #HalfHeartedFanatic

It was a complicated day for logistics. Note the motorcycle helmet in the bike bag in the back of the taxi with my folding bike.

Honestly, I don't know how some car-free and low-car people get by without folding bikes.

#Africa #Madagascar #Antananarivo #Malagasy #Brompton #Multimodality #Carfree #LowCar #Bike #Cycling #CyclingLife #CommuteByBike #NotJustBikes #Urbanism #Photography #OriginalPhoto #HalfHeartedFanatic

SLAY-ASR, или как я перестал волноваться и полюбил тренировать модели

Как добавить аудио-модальность в LLMку максимально экономно? Рассказываю про серию попыток добиться совместимости эмбеддингов разной природы Погрузиться

https://habr.com/ru/articles/1009614/

#representation_learning #multimodality #multimodal_llm #machine_learning #audiomodality #regularization #contrastive_learning #whisper #gemma3

Thrilled to share that Hayk Azizbekyan delivered a compelling presentation on 3D Mixed and Augmented Reality (MR/AR) with AI in Archaeological Fieldwork at the Hong Kong Association for Digital Humanities 2026 Conference!
Taking the stage in the #Image Synthesis, #Immersion & #Multimodality session, Hayk explored how cutting-edge AI-powered 3D MR/AR technologies are revolutionizing archaeological field research—blending digital innovation with cultural heritage exploration and pushing the boundaries of how we study and preserve our past.

This 3-day conference (23-25 Jan) at @cuhkofficial brought together global scholars to explore AI, multimodality, and digital humanities across diverse fields, offering an ideal platform for this groundbreaking research share.

Kudos to Hayk for his excellent work in bridging tech and archaeology!

Details of the conference: https://2026.hkadh.org/

Yue’s research interests lie in natural language processing and information retrieval. Her recent work focuses on LLM-powered AI agents, on advancing the factualness, faithfulness, and trustworthiness of large language models, and on developing and analysing multimodal foundation models.

🔗 Yue’s Webpage: https://fengyue-leah.github.io/

#UKPLab #NLP #InformationRetrieval #LargeLanguageModels #AIResearch #Multimodality #PeerReview #AISafety #TrustworthyAI

Yue’s research interests lie in natural language processing and information retrieval. Her recent work focuses on LLM-powered AI agents, on advancing the factualness, faithfulness, and trustworthiness of large language models, and on developing and analysing multimodal foundation models.

🔗 Yue’s Webpage: https://fengyue-leah.github.io/

#UKPLab #NLP #InformationRetrieval #LargeLanguageModels #AIResearch #Multimodality #PeerReview #AISafety #TrustworthyAI

Kandinsky 5.0: линейка open-source моделей генерации изображений и видео нового поколения

В сентябре мы открыли доступ к Kandinsky 5.0 Video Lite, лёгкой модели с 2 млрд параметров для создания видео по тексту или на основе изображения. Теперь мы выкладываем в open source и все остальные модели линейки Kandinsky 5.0: Video Pro и Image Lite!

https://habr.com/ru/companies/sberbank/articles/971154/

#генерация_видео #kandinsky #multimodality #sberai #sberdevices #generative_models #kandinsky_5 #airi #texttoimage #texttovideo

👻 Probiert es mal aus, egal ob gruselig oder heilig, der „Geist” erwartet euch (es funktionieren auch andere Suchbegriffe, Notizen oder Bilder 😊 )

🧠 P.S. Für alle #CultureKnowledgeGraph- und #SPARQL-Kenner unter euch, bitte hier entlang 👉 https://nfdi4culture.de/resources/knowledge-graph.html

#NFDIrocks #CultureKnowledgeGraph #CultureDataSearch #Multimodality #CulturalHeritage @fiz_karlsruhe
Tabea Tietz @tabea Etienne Posthumus @epoz Harald Sack @lysander07
Linnaea Söhn @linnaea
Torsten Schrade

^ks (^zs)

3/3

👻 Probiert es mal aus, egal ob gruselig oder heilig, der „Geist” erwartet euch (es funktionieren auch andere Suchbegriffe, Notizen oder Bilder 😊 )

🧠 P.S. Für alle #CultureKnowledgeGraph- und #SPARQL-Kenner unter euch, bitte hier entlang 👉 https://nfdi4culture.de/resources/knowledge-graph.html

#NFDIrocks #CultureKnowledgeGraph #CultureDataSearch #Multimodality #CulturalHeritage @fiz_karlsruhe
Tabea Tietz @tabea Etienne Posthumus @epoz Harald Sack @lysander07
Linnaea Söhn @linnaea
Torsten Schrade

^ks (^zs)

3/3

#AndrejKarpathy believes #AGI is still a decade away, citing the need for advancements in #continuallearning, #multimodality, and #computeruse. He argues that while the problems are solvable, they remain challenging. Karpathy also reflects on the history of AI, highlighting the impact of #deeplearning, #reinforcementlearning, and #largelanguagemodels on the field. https://www.dwarkesh.com/p/andrej-karpathy?eicker.news #tech #media #news

#AndrejKarpathy believes #AGI is still a decade away, citing the need for advancements in #continuallearning, #multimodality, and #computeruse. He argues that while the problems are solvable, they remain challenging. Karpathy also reflects on the history of AI, highlighting the impact of #deeplearning, #reinforcementlearning, and #largelanguagemodels on the field. https://www.dwarkesh.com/p/andrej-karpathy?eicker.news #tech #media #news

Как мы учили Алису видеть мир с помощью мультимодальной нейросети Яндекса

Недавно пользователям приложения «Алиса» стал доступен Live-режим, который работает на базе мультимодальной нейросети (VLM), созданной в Яндексе. В этом режиме Алиса распознаёт объекты, показанные ей через камеру смартфона, и рассказывает о них пользователю. А ещё раньше наша VLM стала применяться в Поиске по картинкам, Умной камере и Нейроэксперте. Всё это время технология не стояла на месте и продолжала совершенствоваться. Пожалуй, пришло время поделиться опытом. На связи Роман Исаченко из команды компьютерного зрения в Яндексе. Сегодня я расскажу, какой путь наша VLM прошла за полгода. А Дарья @dara-orange Виноградова, которая работает со мной в той же команде, поделится описанием пайплайна зрения в Алисе. Мы опишем весь путь формирования новой модели: от архитектуры и сбора данных до финальных замеров качества и скорости.

https://habr.com/ru/companies/yandex/articles/904584/

#vlm #natural_language_processing #computer_vision #multimodality #яндекс

“With many genres, one needs the physical impact of the waveforms hitting the body to evaluate what is really going on” —Bob Ludwig, audio mastering engineer. #PhilPerception #AuditoryPerception #multimodality www.soundonsound.com/techniques/m...

Mixing On Headphones

[Перевод] Spring AI научился видеть! Показываю, как заставить GPT находить бананы на картинках

Новый перевод от команды Spring АйО расскажет вам о работе с мультимодальностью при работе со Spring AI и о различных вариантах работы с графическими изображениями с использованием искусственного интеллекта.

https://habr.com/ru/companies/spring_aio/articles/889974/

#spring_ai #multimodality #images #image_generation #image_recognition

Trae: AI Code Editor from ByteDance
https://www.trae.ai
#ycombinator #Trae #Builder #IDE #automatic #collaboration #Multimodality #efficiency #Fast #Accurate #Precise #Context #free #Deploy #editor #code #coding #Autocomplete

Trae: AI Code Editor from ByteDance
https://www.trae.ai
#ycombinator #Trae #Builder #IDE #automatic #collaboration #Multimodality #efficiency #Fast #Accurate #Precise #Context #free #Deploy #editor #code #coding #Autocomplete

Trae: An AI Powered IDE by ByteDance
https://www.trae.ai/home
#ycombinator #Trae #Builder #IDE #automatic #collaboration #Multimodality #efficiency #Fast #Accurate #Precise #Context #free #Deploy #editor #code #coding #Autocomplete

Trae: An AI Powered IDE by ByteDance
https://www.trae.ai/home
#ycombinator #Trae #Builder #IDE #automatic #collaboration #Multimodality #efficiency #Fast #Accurate #Precise #Context #free #Deploy #editor #code #coding #Autocomplete

I am rather happy to notice that @kielipankki, The Language Bank of Finland, is here.

They provide access to numerous resources related to #research, especially relevant to for #linguistics #multimodality #LiteraryStudies and #NaturalLanguageProcessing. Many of the resources are open-access or free to use for #academic pursuits. Check them out!

I am rather happy to notice that @kielipankki, The Language Bank of Finland, is here.

They provide access to numerous resources related to #research, especially relevant to for #linguistics #multimodality #LiteraryStudies and #NaturalLanguageProcessing. Many of the resources are open-access or free to use for #academic pursuits. Check them out!

Kandinsky 4.0 — новая модель генерации видео

Сегодняшний релиз хочется начать с небольшой истории семейства моделей Kandinsky. В прошлом году на конференции AI Journey 2023 наша команда представила две модели: Kandinsky 3.0 для генерации изображений и первую российскую модель генерации видео по тексту Kandinsky Video . В этом году в апреле и мае вышли в свет улучшенные версии этих моделей: Kandinsky 3.1 с повышенным качеством изображений и Kandinsky Video 1.1 с улучшенными визуальным качеством и временной связностью кадров на видео. С тех пор прогресс в области генеративных моделей привел к созданию множества интересных решений для задач генерации, связывающих текст, видео и аудио модальности. Сегодня наша команда исследователей и учёных из Лаборатории Sber AI Research при поддержке учёных Лаборатории FusionBrain Института AIRI представляет Kandinsky 4.0 — нейросеть нового поколения для создания реалистичных видеороликов по текстовому описанию или стартовому кадру, а также аудио сопровождения для сгенерированного видеоролика. Теперь модель генерирует видеоряд продолжительностью до 12 секунд в разрешении HD (1280x720) по любому текстовому описанию или произвольному стартовому кадру. С помощью модели можно создавать видео с разным соотношением сторон под любые пользовательские и продуктовые потребности. В этой статье мы подробно рассмотрим структуру, функционал и процесс обучения нашей новой модели.

https://habr.com/ru/companies/sberbank/articles/866156/

#генерация_видео #kandinsky #video #multimodality #sberai #sberdevices #airi #generative_models #kandinsky_4

Gleich geht es los mit der Panel Discussion zum Thema "Wissen durch Bilder", am
@dipf_aktuell, @bbf_dipfberlin. Ich freue mich auf den Austausch zu Bildern, Texten, multimodalen Formaten. #Bildungsgeschichte und #DigitalHumanities im Dialog! #DH #Multimodality

Gleich geht es los mit der Panel Discussion zum Thema "Wissen durch Bilder", am
@dipf_aktuell, @bbf_dipfberlin. Ich freue mich auf den Austausch zu Bildern, Texten, multimodalen Formaten. #Bildungsgeschichte und #DigitalHumanities im Dialog! #DH #Multimodality

I am heading to #Vienna for taking part at the Hackathon on #Multimodality in Digital Humanities. I am extremely glad, that years ago I took part at a three-days workshop on analysing DEMs for archaeology and one of the participants was 60y. That inspired me, to never stop visiting hackathons, workshops and the like, as they are unique opportunities to learn and to tackle my ideas. #motivatedgranny👵🏼😊

I am heading to #Vienna for taking part at the Hackathon on #Multimodality in Digital Humanities. I am extremely glad, that years ago I took part at a three-days workshop on analysing DEMs for archaeology and one of the participants was 60y. That inspired me, to never stop visiting hackathons, workshops and the like, as they are unique opportunities to learn and to tackle my ideas. #motivatedgranny👵🏼😊

Small-scale accessible transport: The Baby Bus. There's a FRONT BIKE RACK!!! – another missing element in many minibus-based transport networks in the global South.

https://youtu.be/QI1yM9mzCAE

#publictransport #sustainabletransport #transportation #Accessibility #Americas #bus #BusRapidTransit #BRT #ElectricVehicles #EV #EVs #multimodal #Multimodality #publictransit #TrufiAssociation

Job Opening: Post-doctoral researcher in Computer Science or Digital Humanities at the University of Helsinki. Join the ERC-funded project on multimodality starting in Jan 2025.
Apply by Oct 21, 2024.
More info: https://jobs.helsinki.fi/job-invite/3513/ @tuomo
#PostDoc #DigitalHumanities #Multimodality

#CfP for the #conference "Femmes et #canon #littéraire à l’ère de la #multimodalité", which will take place at the Université de Bourgogne (Dijon) on November 7-8, 2024.

🗓️Deadline for Abstracts: August 15, 2024

📌Further Information: https://www.avldigital.de/de/vernetzen/details/callforpapers/femmes-et-canon-litteraire-a-lere-de-la-multimodalite-dijon/ #LiteraryHistory #LiteratureHistory #LiteratureGender #Intermediality #multimodality

I’ve been playing around with them occasionally and I just can’t get them to produce what I’m actually asking for. For example this is Luma’s response to the prompt “A deeply content capybara luxuriating in my suburban back garden with a fluffy black cat reclining on his belly“:

Is this a capybara or is it a cross between a pig and a warthog? Where’s the black cat? Perhaps it’s a bad prompt but it’s a pattern I’ve noticed across these systems. This is Runway’s response to the prompt “We want to make a video about a capybara in a tuxedo singing a song about his life, while holding an umbrella because it’s raining. His capybara wife has got tears in her eyes as she hears him sing because his singing is terrible. It hurts her ears. remember the umbrella and sound” (I’m doing this with kids in case you’re wondering…)

On level stunning. On another more practical level, kind of useless. I’ve yet to put any significant thought into how I’m prompting video systems, so I suspect there might at least some user failure here. But I’m confident there’s a broader weakness in the reliability of video systems because you just can’t infer robustly for multimodal content in the way you can for text.

Will these always be superficially spectacular but deeply useless in practice? If so is a lot of money being wasted? Or do we risk a future in which a much smaller number of designers are employed to clean up after chronically dysfunctional video models?

https://markcarrigan.net/2024/07/26/the-ai-video-generators-are-superficially-stunning-but-deeply-useless/

#generativeAI #Luma #multimodality #Runway #video

Challenge at the AI - geodata interface! 31.07.24 Online event "HEGEMON" of the @CybAgBund. Registration: https://t1p.de/m9vpi
#Research #Innovation #FoundationModels #GenerativeAI #Benchmarking #LLM #Multimodality #Geodata #GIS
https://nachrichten.idw-online.de/2024/07/02/press-release-cyberagentur-analysing-foundation-models-in-the-security-context

I write about #collaboration and #multimodality. That includes multisensory experiences - and now includes multisensory branding. https://www.axios.com/2024/06/14/multisensory-marketing-mastercard-branding-priceless

Wow! 🥳🎊 I received a copy of a book we wrote a chapter for with @tuomo and Hanna-Mari Pienimäki.

It's one thing to see one's own work published on a website, but another feeling when you can hold it in your hand.

If you're interested in #multilingualism #sociolinguistics #UrbanStudies and #LinguisticLandscapes check out this book. Relevant also for #UrbanGeography #HumanGeography #GIScience and #multimodality 🤓

The #OpenAccess book is available here: https://oa.finlit.fi/site/books/e/10.21435/sflin.24/

A warm welcome to Falko Helm, who has just started as a PhD candidate at the UKP Lab! 👋 Falko researches #Multimodality & Structure for #TransformerModels. He is also interested in #GraphTheory to handle linked documents. Find more about Falko here: https://github.com/Falko1