Idil (@idilsulo) header
Idil (@idilsulo) profile picture

Researcher in computer vision | 🎓 MSc @TU_Muenchen CS, BSc @METU_ODTU CENG

Berlin
Joined January 2018
Photo by Idil (@idilsulo)Photo by Idil (@idilsulo)Photo by Idil (@idilsulo)Photo by Idil (@idilsulo)Photo by Idil (@idilsulo)Photo by Idil (@idilsulo)Photo by Idil (@idilsulo)Photo by Idil (@idilsulo)Photo by Idil (@idilsulo)Photo by Idil (@idilsulo)
Home Robotics'te Beko ile işbirliğinin eşiğindeyiz. Lisans ve lisansüstü araştırmacılar arıyoruz. İlgilenenlerin bir an önce başvurmalarını (bazı ufak görevlerin yapılmasını bekleyeceğiz) öneriyorum. Not: herkese ulaştırmak için takipçilerimin desteğini rica ediyorum.
8
22
4,451
If you're in Milan at #ECCV2024, check out our work #SceneTeller on high-quality 3D scene generation from natural language prompts! 🏡 Project page: sceneteller.github.io/ 📄 Paper: arxiv.org/abs/2407.20727 👩‍💻 Code: github.com/sceneteller/Scene… @eccvconf
2
11
62
4,995
Introducing The AI Scientist: The world’s first AI system for automating scientific research and open-ended discovery! sakana.ai/ai-scientist/ From ideation, writing code, running experiments and summarizing results, to writing entire papers and conducting peer-review, The AI Scientist opens a new era of AI-driven scientific research and accelerated discovery. Here are 4 example Machine Learning research papers generated by The AI Scientist. We published our report, The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery, and open-sourced our project! Paper: arxiv.org/abs/2408.06292 GitHub: github.com/SakanaAI/AI-Scien… Our system leverages LLMs to propose and implement new research directions. Here, we first apply The AI Scientist to conduct Machine Learning research. Crucially, our system is capable of executing the entire ML research lifecycle: from inventing research ideas and experiments, writing code, to executing experiments on GPUs and gathering results. It can also write an entire scientific paper, explaining, visualizing and contextualizing the results. Furthermore, while an LLM author writes entire research papers, another LLM reviewer critiques resulting manuscripts to provide feedback to improve the work, and also to select the most promising ideas to further develop in the next iteration cycle, leading to continual, open-ended discoveries, thus emulating the human scientific community. As a proof of concept, our system produced papers with novel contributions in ML research domains such language modeling, Diffusion and Grokking. We (@_chris_lu_, @RobertTLange, @hardmaru) proudly collaborated with the @UniOfOxford (@j_foerst, @FLAIR_Ox) and @UBC (@cong_ml, @jeffclune) on this exciting project.
282
1,455
727
6,001
3,433,022
82
7
859
51,668
$5 for a bagel? bro that’s 3 full hours of a100 on the cloud
2
3
87
5,541
What is 3D computer vision? #3DV2024
7
79
11
488
68,831
⚠️ Confirmed: Real-time network data show Twitter has been restricted in #Turkey; the filtering is applied on major internet providers and comes as the public come to rely on the service in the aftermath of a series of deadly earthquakes 📰 Report: netblocks.org/reports/twitte…
412
6,145
3,683
10,473
7,342,761
Enkaz kaldırma çok yetersiz. Sokaktan cesetler bile toplanmıyor
802
9,371
602
37,845
4,025,304
I'm getting asked how to help Türkiye 🇹🇷internationally I've gathered some NGOs accepting funds from different countries and crypto addresses. Please let me know of any other resources you are aware of earthquaketurkey.notion.site…
4
25
4
56
16,055
The war on children intensifies.
1,183
2,758
937
36,537
13,541,916
Me: Nobody understands my sense of humor AI: I got you.
New large language model released by @GoogleAI. Take a look, it's pretty hard to believe. It can explain jokes:
17
121
41
741
Meeting the rebuttal deadline
2
38
8
348
Building the first instance of a thing is the hardest part. This makes me appreciative to live in a world of brilliant humans whose "shoulders" we can stand on, but also it's motivating me to try to build a thing that doesn't yet exist and might help at least one other person.
114
667
36
6,049
85
4,264
181
12,502
Pretty amazing adversarial image for the human visual system. Outside the fovea, feature detectors are sensitive to lower-spatial frequency signals and pooled over large areas. One loses the high-frequency forest for the low-frequency tree.
This deceiving grid tricks you into thinking there's a curved line somewhere, but you can't find it. The purposefully placed gray lines of squares in a curved formation will induce your peripheral vision to interpolate curved lines [read more: buff.ly/3oKqWXa]
9
155
7
836
Can you imagine if a self-driving car saw this
251
9,022
357
118,327
'Firefighters are being completely outpaced by the flames and the fire.' Our special correspondent @AlexCrawfordSky reports from Bodrum in Turkey where wildfires are burning out of control. Read more here: trib.al/3SVnNsr
337
5,288
1,400
14,926
Umut bitmez.
428
969
65
24,801
“Today, everybody is talking about self-driving cars. Such truly self-driving cars first appeared in the 1980s-90s in Munich…” #GTC21
18
38
1
350
Turkey: Bogazici faculty today, after police detained 159 students yesterday.
122
4,821
271
41,627