
Pritam
@pritam94
Followers
163
Following
745
Media
21
Statuses
183
PhD candidate at @queensu and affiliate at @VectorInst. Ex: @Google, @BorealisAI, @Infosys, @tech_mahindra. opinions are my own.
Kingston, Ontario
Joined April 2011
๐ Excited to share that my @Google internship work has been accepted at @iclr_conf #ICLR2025! ๐. ๐ Paper: ๐งต.
2
3
21
RT @phillip_isola: Our computer vision textbook is now available for free online here:. We are working on adding soโฆ.
visionbook.mit.edu
0
620
0
I am attending #ICLR2025 to present our paper โPhrase-Level Alignment to Mitigate Vision-Language Hallucinations.โ c.Come check out our poster! Looking forward to meeting some old friends and making some new ones. Happy to chat about #multimodal models & #video understanding
0
0
2
very well done!.
#CVPR2025 Area Chairs (ACs) identified a number of highly irresponsible reviewers, those who either abandoned the review process entirely or submitted egregiously low-quality reviews, including some generated by large language models (LLMs). 1/2.
0
0
0
๐ข Also, ๐ ๐ฎ๐บ ๐น๐ผ๐ผ๐ธ๐ถ๐ป๐ด ๐ณ๐ผ๐ฟ ๐ฎ ๐ณ๐๐น๐น-๐๐ถ๐บ๐ฒ ๐ฟ๐ฒ๐๐ฒ๐ฎ๐ฟ๐ฐ๐ต๐ฒ๐ฟ ๐ฝ๐ผ๐๐ถ๐๐ถ๐ผ๐ปโbroadly in the areas of video and multimodal learning. Please donโt hesitate to reach out if you find my experience a good fit. ๐ข. ๐บ๐ผ๐ฟ๐ฒ ๐ฎ๐ฏ๐ผ๐๐ ๐บ๐ฒ:
0
0
1
A huge shoutout to my amazing co-authors: ๐๐ @SaynaEbrahimi @Ali_Etemad1 @abeirami @sercanarik @tomaspfister.
1
0
2
๐ง๐;๐๐ฅ: Existing hallucination mitigation methods often hurt the general capabilities of multimodal LLMs. We introduce ๐๐ฎ๐๐ฎ-๐ฎ๐๐ด๐บ๐ฒ๐ป๐๐ฒ๐ฑ ๐ฃ๐ต๐ฟ๐ฎ๐๐ฒ-๐น๐ฒ๐๐ฒ๐น ๐๐น๐ถ๐ด๐ป๐บ๐ฒ๐ป๐, a fine-grained mechanism to mitigate hallucinations w/o hurting general capabilities.
1
0
2
RT @iclr_conf: Financial Assistance applications are now open! If you face financial barriers to attending ICLR 2025, we encourage you to aโฆ.
0
28
0
RT @NobelPrize: Watch the Nobel Prize lectures in physics. See 2024 physics laureates John Hopfield and Geoffrey Hinton deliver their Nโฆ.
0
217
0
RT @iclr_conf: โ๏ธ Reminder to reviewers: Check author responses to your reviews, and ask follow up questions if needed. 50% of papers haveโฆ.
0
24
0
RT @iclr_conf: ๐ฃ๏ธ ICLR response period ends soon -- remember to check author responses, ask follow-up questions, and revise scores. Revieโฆ.
0
21
0
RT @jbhuang0604: The more I learn vision-language models, the more I realize how little *vision* there is in them. ๐ค. Model: Image as a seqโฆ.
0
51
0
RT @NobelPrize: BREAKING NEWS.The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Chemistry with one half toโฆ.
0
9K
0