PinnedRitvik RastogiThanks for the appreciation, Its surreal for me to get acknowledged from the author itself.Feb 1Feb 1
Ritvik RastogiPapers Explained 159: XLM RobertaXLM-RoBERTa combines RoBERTa techniques with XLM, excluding translation language modelling. Instead, it focuses on masked language…2d ago2d ago
Ritvik RastogiPapers Explained 158: XLMXLM is a transformer-based model, built by Meta. It extends the approach of generative pretraining to multiple languages and shows the…4d ago4d ago
Ritvik RastogiPapers Explained 157: Gemma 2Gemma 2 is a new addition to the Gemma family with several technical modifications, including interleaving local-global attentions and…6d ago6d ago
Ritvik RastogiPapers Explained 156: InstructBLIPThis paper conducts a systematic and comprehensive study on vision-language instruction tuning based on the pretrained BLIP-2 models. 26…Jun 28Jun 28
Ritvik RastogiPapers Explained 155: BLIP 2BLIP-2 is a generic and efficient pretraining strategy that bootstraps vision-language pre-training from off-the-shelf frozen pre-trained…Jun 26Jun 26
Ritvik RastogiPapers Explained 154: BLIPBLIP is a new VLP framework which transfers flexibly to both vision-language understanding and generation tasks. BLIP effectively utilizes…Jun 24Jun 24
Ritvik RastogiPapers Explained 153: CTRLCTRL is a 1.63 billion-parameter conditional transformer language model, trained to condition on control codes that govern style, content…Jun 21Jun 21
Ritvik RastogiPapers Explained 152: SigLipThis paper proposes a simple pairwise Sigmoid loss for Language-Image Pre-training (SigLIP). Unlike standard contrastive learning with…Jun 20Jun 20
Ritvik RastogiPapers Explained 151: Aya 23Aya 23 is a family of multilingual language models that can serve 23 languages. It is an improvement over the previous model, Aya 101…Jun 17Jun 17