Kartik Thakral
@thakralkartik78
Followers
55
Following
2K
Media
1
Statuses
86
Misfit
Jodhpur, India
Joined June 2020
Our ICML paper on "Harmonizing Geometry and Uncertainty: Diffusion with Hyperspheres" is available at arxiv: https://t.co/AKttP2Dm59 With Muskan, Chiranjeev, @thakralkartik78, and @RichaSingh26
#diffusionmodel #diffusion #genAI #IndiaAI #deeplearning #ICML2025 #generativeAI
arxiv.org
Do contemporary diffusion models preserve the class geometry of hyperspherical data? Standard diffusion models rely on isotropic Gaussian noise in the forward process, inherently favoring...
0
2
18
There is a week left for submissions to the 2nd edition of the Unlearning & Model Editing (U&Me) workshop at #ICCV2025. We are accepting both long (archival) & short(non-archival) submissions. Last year's U&Me workshop with @davidbau & @rita_cucchiara at #ECCV was a joy & ...
1
2
2
š„ News - 2nd Unlearning and Model Editing Workshop and Challenge at #ICCV2025 š Call for papers ready and OpenReview accepting submissions: https://t.co/1zM0M21Blv š§© New challenge on #Unlearning
https://t.co/QodoHDwJcp / https://t.co/K6lIBwEBtz Best performers in paper!
0
7
11
"Harmonizing Geometry and Uncertainty: Diffusion with Hyperspheres" accepted in #ICML20025 jointly with Muskan, Chiranjeev, Kartik, and @RichaSingh26
#machinelearning #diffusionmodel #diffusion #generativeAI #artificialintelligence #computervision #genAI paper coming soon
0
2
15
In Singapore, presenting our deepfake research at #ICLR2025 with @thakralkartik78, Akanksha Singh, @RichaSingh26
#deeplearning #genAI #deepfake #misinformation #iitjodhpur
1
4
30
2nd edition of U&ME workshop will be atĀ #ICCV2025. After the success of 1st workshop atĀ #ECCV2024, weāre gearing up for an even more exciting programāincluding our first-ever competitionĀ ... details coming soon With @_iAc @dgolano @THassner @thakralkartik78
#Unlearning #genAI
0
4
13
āILLUSION: Unveiling Truth with a Comprehensive Multi-Modal, Multi-Lingual Deepfake Datasetā has been accepted at #ICLR2025 ... it is a large-scale dataset of 1.3+ million multi-modal, multi-lingual (26+ languages) audio-video #deepfake samples. #SyntheticMedia #responsibleAI
0
6
25
š Unmasking (Truly) Deepfakes šµļøāāļø We are organising IEEE CAI 2025 Workshop to tackle deepfake challenges across all modalitiesānot just videos! šļø Submission Deadline: Jan 15, 2025 š Details: https://t.co/U3llzKJPP6
#Deepfakes #AI #CAI2025 #ResponsibleAI #Misinformation
sites.google.com
Final Paper Submission March 10, 2025, 23:59:59 PDT (Select the Workshop on Unmasking (Truly) Deepfakes: Not only Video Deepfakes Track to submit the paper in this workshop)
0
2
4
A slightly late post but better late than never. During the week of October 22nd, India welcomed two leaders of the AI world: Prof. @ylecun (Meta/NYU) and Mr. Jensen Huang (CEO of NVIDIA). The NVIDIA AI Summit was nothing short of phenomenal, with over 7,000 attendees. I had
1
5
24
October Update! š 1 paper accepted to #WACV2025 with @NithinGK10 @vishalm_patel 1 paper accepted to IEEE T-BIOM with @thakralkartik78 @surbhimittal97 @MayankVatsa3 @RichaSingh26 Stay tuned for the paper and code! #WACV #TBIOM #IEEE @wacv_official
1
1
9
Deeply honored to be recognized with the Shakti Mentor Award by the Google News Initiative! š This recognition stands as a testament to our team's dedication and innovative work in advancing deepfake detection technologies. Together, we are pushing the boundaries of AI to
1
5
25
Our latest research, "On responsible machine learning datasets emphasizing fairness, privacy, and regulatory norms with examples in biometrics and healthcare," has been published in Nature Machine Intelligence (Impact Factor 18.8)! We conducted a large-scale audit of biometric
2
3
19
What an incredible evening for our lab! š We are thrilled to announce that we received the prestigious @nasscom AI Gamechangers 2023-24 awards for our research on face recognition and deepfakes. The awards were announced at the @nasscomai Confluence, and it was an honor for
2
3
38
#CODSCOMAD December 2024 edition will be held at IIT Jodhpur. The call for papers is available at https://t.co/0phYK6DoJk We are also looking to expand our pool of reviewers and invite researchers to not only submit their papers but also nominate themselves as reviewers. If
0
5
21
The Department of Computer Science & Engineering @ IIT Jodhpur celebrates ECCV 24 publications!! #iitjodhpur #cseiitjodhpur #eccv2024 #happytimes
0
3
4
Excited to share that two papers are accepted at #eccv2024! Paper 1 (with @Chiranj18308959, @Muskan_Dosi, @thakralkartik78, @RichaSingh26): HyperSpaceX: Radial and Angular Exploration of HyperSpherical Dimensions proposes a new method for image classification & recognition that
1
5
23
Our research on #GenAI (deepfake detection) was highlighted in the episode "Facing AI's Evolution! AI Gets Personal | Becoming Human 4" starting at the 20:45 mark. We had the pleasure of hosting @chuaenlai and the @ChannelNewsAsia team at the #IITJodhpur campus.
0
3
15
Way to go Dr. Pupu!
Puspita successfully defended her dissertation. Facial analysis and recognition technologies often exhibit biased and unreliable performance, especially when dealing with diverse demographic groups and altered facial conditions, such as injuries. Puspita's research addresses
0
0
2