Violeta Chang (@vnchangc) 's Twitter Profile
Violeta Chang

@vnchangc

Researcher and professor at @diinf @usach
Artificial intelligence with social impact

ID: 2617518168

calendar_today11-07-2014 15:16:16

146 Tweet

108 Followers

193 Following

The Clinic (@thecliniccl) 's Twitter Profile Photo

Especialista señala que un celular con muchas notificaciones y desorden digital puede ser un síntoma de un problema para la salud mental. theclinic.cl/2024/11/16/des…

khipu.ai (@khipu_ai) 's Twitter Profile Photo

📣 A pocos días del comienzo del Khipu 2025, nos complace anunciar que tanto las actividades del salón principal como el acto de clausura del viernes se retransmitirán en directo por este canal: khipu.ai/live/. ¡Los esperamos!

Violeta Chang (@vnchangc) 's Twitter Profile Photo

Ayer comenzó #khipu2025 en Santiago de Chile!! Un evento que trae lo mejor de IA para LATAM. La primera jornada terminó con un taller para hacer nuestro propio Quipu con la fecha de nacimiento, después de escuchar maravillosas charlas sobre IA.

Ayer comenzó #khipu2025 en Santiago de Chile!! Un evento que trae lo mejor de IA para LATAM. La primera jornada terminó con un taller para hacer nuestro propio Quipu con la fecha de nacimiento, después de escuchar maravillosas charlas sobre IA.
Violeta Chang (@vnchangc) 's Twitter Profile Photo

What an amazing journey at #khipu2025 with lectures on Computer Vision by Leo Sampaio, Graph Neural Networks by petar velickovic and Diffusion Models by Ira Ktena. Very happy to share khipu.ai with my students from Informática USACH Ingeniería USACH

What an amazing journey at #khipu2025 with lectures on Computer Vision by Leo Sampaio, Graph Neural Networks by <a href="/petarvelickovic/">petar velickovic</a> and Diffusion Models by Ira Ktena. Very happy to share <a href="/Khipu_AI/">khipu.ai</a> with my students from <a href="/diinf/">Informática USACH</a> <a href="/FingUsach/">Ingeniería USACH</a>
Violeta Chang (@vnchangc) 's Twitter Profile Photo

Fantastic last day at khipu.ai with outstanding talks for a broader audience. First, we had a Women in AI event. Then, the closing event included inspiring talks by Jeff Dean and Joelle Pineau, among others. A very nice end to #KHIPU2025 at Centro Cultural CEINA

Fantastic last day at <a href="/Khipu_AI/">khipu.ai</a> with outstanding talks for a broader audience. First, we had a Women in AI event. Then, the closing event included inspiring talks by <a href="/JeffDean/">Jeff Dean</a> and <a href="/jpineau1/">Joelle Pineau</a>, among others. 
A very nice end to #KHIPU2025 at <a href="/Centro_Ceina/">Centro Cultural CEINA</a>
Violeta Chang (@vnchangc) 's Twitter Profile Photo

Having had khipu.ai in Santiago de Chile is indescribable 🥰 #KHIPU2025 was possible thanks to a great volunteer work of an excellent team of researchers. I hope all the participants enjoyed it as much as I did 😍 See you at #KHIPU2027!!

Having had <a href="/Khipu_AI/">khipu.ai</a> in Santiago de Chile is indescribable 🥰 
#KHIPU2025 was possible thanks to a great volunteer work of an excellent team of researchers. I hope all the participants enjoyed it as much as I did 😍
See you at #KHIPU2027!!
Tanishq Mathew Abraham, Ph.D. (@iscienceluvr) 's Twitter Profile Photo

Perception Encoder: The best visual embeddings are not at the output of the network "we find that contrastive vision-language training alone can produce strong, general embeddings for all of these downstream tasks. There is only one caveat: these embeddings are hidden within the

Perception Encoder: The best visual embeddings are not at the output of the network

"we find that contrastive vision-language training alone can produce strong, general embeddings for all of these downstream tasks. There is only one caveat: these embeddings are hidden within the
Sara Hooker (@sarahookr) 's Twitter Profile Photo

It is critical for scientific integrity that we trust our measure of progress. The lmarena.ai has become the go-to evaluation for AI progress. Our release today demonstrates the difficulty in maintaining fair evaluations on lmarena.ai, despite best intentions.

It is critical for scientific integrity that we trust our measure of progress. 

The <a href="/lmarena_ai/">lmarena.ai</a> has become the go-to evaluation for AI progress.

Our release today demonstrates the difficulty in maintaining fair evaluations on <a href="/lmarena_ai/">lmarena.ai</a>, despite best intentions.