1 code implementation • 9 Oct 2024 • Abhinav Shukla, Sai Vemprala, Aditya Kusupati, Ashish Kapoor
In this work, we present MatMamba: a state space model which combines Matryoshka-style learning with Mamba2, by modifying the block to contain nested dimensions to enable joint training and adaptive inference.
no code implementations • 8 Oct 2024 • Mohammadreza Salehi, Jae Sung Park, Tanush Yadav, Aditya Kusupati, Ranjay Krishna, Yejin Choi, Hannaneh Hajishirzi, Ali Farhadi
To evaluate the effectiveness of multimodal foundation models in helping us recognize such actions, we present ActionAtlas v1. 0, a multiple-choice video question answering benchmark featuring short videos across various sports.
no code implementations • 29 Jul 2024 • Gagan Jain, Nidhi Hegde, Aditya Kusupati, Arsha Nagrani, Shyamal Buch, Prateek Jain, Anurag Arnab, Sujoy Paul
We present Mixture of Nested Experts (MoNE), which utilizes a nested structure for experts, wherein individual experts fall on an increasing compute-accuracy curve.
1 code implementation • 28 May 2024 • Ethan Shen, Alan Fan, Sarah M. Pratt, Jae Sung Park, Matthew Wallingford, Sham M. Kakade, Ari Holtzman, Ranjay Krishna, Ali Farhadi, Aditya Kusupati
We achieve this by feeding a superposition of the most recent token embeddings from the $k$ drafts as input to the next decoding step of the language model.
no code implementations • 29 Mar 2024 • Jinhyuk Lee, Zhuyun Dai, Xiaoqi Ren, Blair Chen, Daniel Cer, Jeremy R. Cole, Kai Hui, Michael Boratko, Rajvi Kapadia, Wen Ding, Yi Luan, Sai Meher Karthik Duddu, Gustavo Hernandez Abrego, Weiqiang Shi, Nithi Gupta, Aditya Kusupati, Prateek Jain, Siddhartha Reddy Jonnalagadda, Ming-Wei Chang, Iftekhar Naim
On the Massive Text Embedding Benchmark (MTEB), Gecko with 256 embedding dimensions outperforms all existing entries with 768 embedding size.
1 code implementation • 9 Nov 2023 • Ethan Shen, Ali Farhadi, Aditya Kusupati
In this work, we set out to investigate if hierarchical visual representations truly capture the human perceived hierarchy better than standard learned representations.
no code implementations • 18 Oct 2023 • Mohammadreza Salehi, Sachin Mehta, Aditya Kusupati, Ali Farhadi, Hannaneh Hajishirzi
We introduce SHARCS for adaptive inference that takes into account the hardness of input samples.
no code implementations • 13 Oct 2023 • Ramnath Kumar, Anshul Mittal, Nilesh Gupta, Aditya Kusupati, Inderjit Dhillon, Prateek Jain
To facilitate the effective learning of this discrete structure, EHI introduces dense path embeddings that encodes the path traversed by queries and documents within the tree.
2 code implementations • 11 Oct 2023 • Devvrit, Sneha Kudugunta, Aditya Kusupati, Tim Dettmers, KaiFeng Chen, Inderjit Dhillon, Yulia Tsvetkov, Hannaneh Hajishirzi, Sham Kakade, Ali Farhadi, Prateek Jain
Furthermore, we observe that smaller encoders extracted from a universal MatFormer-based ViT (MatViT) encoder preserve the metric-space structure for adaptive large-scale retrieval.
1 code implementation • NeurIPS 2023 • Matthew Wallingford, Vivek Ramanujan, Alex Fang, Aditya Kusupati, Roozbeh Mottaghi, Aniruddha Kembhavi, Ludwig Schmidt, Ali Farhadi
Performing lightweight updates on the recalled data significantly improves accuracy across a variety of distribution shift and transfer learning benchmarks.
1 code implementation • NeurIPS 2023 • Aniket Rege, Aditya Kusupati, Sharan Ranjit S, Alan Fan, Qingqing Cao, Sham Kakade, Prateek Jain, Ali Farhadi
Finally, we demonstrate that AdANNS can enable inference-time adaptivity for compute-aware search on ANNS indices built non-adaptively on matryoshka representations.
1 code implementation • 10 Jan 2023 • Matthew Wallingford, Aditya Kusupati, Alex Fang, Vivek Ramanujan, Aniruddha Kembhavi, Roozbeh Mottaghi, Ali Farhadi
Compositional representations of the world are a promising step towards enabling high-level scene understanding and efficient transfer to downstream tasks.
4 code implementations • 26 May 2022 • Aditya Kusupati, Gantavya Bhatt, Aniket Rege, Matthew Wallingford, Aditya Sinha, Vivek Ramanujan, William Howard-Snyder, KaiFeng Chen, Sham Kakade, Prateek Jain, Ali Farhadi
The flexibility within the learned Matryoshka Representations offer: (a) up to 14x smaller embedding size for ImageNet-1K classification at the same level of accuracy; (b) up to 14x real-world speed-ups for large-scale retrieval on ImageNet-1K and 4K; and (c) up to 2% accuracy improvements for long-tail few-shot classification, all while being as robust as the original representations.
Ranked #25 on Image Classification on ObjectNet (using extra training data)
no code implementations • 22 Feb 2022 • Dhruv Jain, Khoa Huynh Anh Nguyen, Steven Goodman, Rachel Grossman-Kahn, Hung Ngo, Aditya Kusupati, Ruofei Du, Alex Olwal, Leah Findlater, Jon E. Froehlich
Recent advances have enabled automatic sound recognition systems for deaf and hard of hearing (DHH) users on mobile devices.
no code implementations • CVPR 2022 • Rowan Zellers, Jiasen Lu, Ximing Lu, Youngjae Yu, Yanpeng Zhao, Mohammadreza Salehi, Aditya Kusupati, Jack Hessel, Ali Farhadi, Yejin Choi
Given a video, we replace snippets of text and audio with a MASK token; the model learns by choosing the correct masked-out snippet.
Ranked #6 on Action Classification on Kinetics-600 (using extra training data)
no code implementations • ICML Workshop AML 2021 • Ivan Evtimov, Ian Covert, Aditya Kusupati, Tadayoshi Kohno
When data is publicly released for human consumption, it is unclear how to prevent its unauthorized usage for machine learning purposes.
1 code implementation • NeurIPS 2021 • Aditya Kusupati, Matthew Wallingford, Vivek Ramanujan, Raghav Somani, Jae Sung Park, Krishna Pillutla, Prateek Jain, Sham Kakade, Ali Farhadi
We further quantitatively measure the quality of our codes by applying it to the efficient image retrieval as well as out-of-distribution (OOD) detection problems.
2 code implementations • 6 Jul 2020 • Matthew Wallingford, Aditya Kusupati, Keivan Alizadeh-Vahid, Aaron Walsman, Aniruddha Kembhavi, Ali Farhadi
To foster research towards the goal of general ML methods, we introduce a new unified evaluation framework - FLUID (Flexible Sequential Data).
2 code implementations • NeurIPS 2020 • Oindrila Saha, Aditya Kusupati, Harsha Vardhan Simhadri, Manik Varma, Prateek Jain
Standard Convolutional Neural Networks (CNNs) designed for computer vision tasks tend to have large intermediate activation maps.
Ranked #26 on Face Detection on WIDER Face (Medium)
1 code implementation • ICML 2020 • Aditya Kusupati, Vivek Ramanujan, Raghav Somani, Mitchell Wortsman, Prateek Jain, Sham Kakade, Ali Farhadi
Sparsity in Deep Neural Networks (DNNs) is studied extensively with the focus of maximizing prediction accuracy given an overall parameter budget.
no code implementations • 15 Jan 2020 • Yashoteja Prabhu, Aditya Kusupati, Nilesh Gupta, Manik Varma
This paper also introduces a (3) new labelwise prediction algorithm in XReg useful for DSA and other recommendation tasks.
1 code implementation • 6 Sep 2019 • Dhrubojyoti Roy, Sangeeta Srivastava, Aditya Kusupati, Pranshu Jain, Manik Varma, Anish Arora
Edge sensing with micro-power pulse-Doppler radars is an emergent domain in monitoring and surveillance with several smart city applications.
1 code implementation • NeurIPS 2018 • Aditya Kusupati, Manish Singh, Kush Bhatia, Ashish Kumar, Prateek Jain, Manik Varma
FastRNN addresses these limitations by adding a residual connection that does not constrain the range of the singular values explicitly and has only two extra scalar parameters.