Darshana SaravananI am Research Intern at Microsoft Research India where I am exploring AI for Science, under the mentorship of Dr. Navin Goyal. I recently completed my MS by Research in Computer Science at IIIT Hyderabad, where I worked on understanding compositionality in Vision Language Models. I was advised by Dr. Vineet Gandhi and had the opportunity to collaborate with Dr. Makarand Tapaswi and Dr. Naresh Manwani. Email / GitHub / Google Scholar / LinkedIn |
|
ResearchI am interested in understanding and improving multimodal AI systems through interpretability. |
Publications |
|
Investigating Mechanisms for In-Context Vision Language BindingDarshana Saravanan, Makarand Tapaswi, Vineet Gandhi Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), MIV, 2025 Oral arxiv / |
|
VELOCITI: Benchmarking Video-Language Compositional Reasoning with Strict EntailmentDarshana Saravanan*, Varun Gupta*, Darshan Singh*, Zeeshan Khan, Vineet Gandhi, Makarand Tapaswi Conference on Computer Vision and Pattern Recognition (CVPR), 2025 arxiv / website / |
|
Pseudo-labelling meets Label Smoothing for Noisy Partial Label LearningDarshana Saravanan, Naresh Manwani, Vineet Gandhi Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), FGVC-12, 2025 Best Paper Award arxiv / code / |
|
Design and source code from Jon Barron's website |