Alert button
Picture for Anshumali Shrivastava

Anshumali Shrivastava

Alert button

CAPS: A Practical Partition Index for Filtered Similarity Search

Aug 29, 2023
Gaurav Gupta, Jonah Yi, Benjamin Coleman, Chen Luo, Vihan Lakshman, Anshumali Shrivastava

Figure 1 for CAPS: A Practical Partition Index for Filtered Similarity Search
Figure 2 for CAPS: A Practical Partition Index for Filtered Similarity Search
Figure 3 for CAPS: A Practical Partition Index for Filtered Similarity Search
Figure 4 for CAPS: A Practical Partition Index for Filtered Similarity Search
Viaarxiv icon

Scissorhands: Exploiting the Persistence of Importance Hypothesis for LLM KV Cache Compression at Test Time

May 26, 2023
Zichang Liu, Aditya Desai, Fangshuo Liao, Weitao Wang, Victor Xie, Zhaozhuo Xu, Anastasios Kyrillidis, Anshumali Shrivastava

Figure 1 for Scissorhands: Exploiting the Persistence of Importance Hypothesis for LLM KV Cache Compression at Test Time
Figure 2 for Scissorhands: Exploiting the Persistence of Importance Hypothesis for LLM KV Cache Compression at Test Time
Figure 3 for Scissorhands: Exploiting the Persistence of Importance Hypothesis for LLM KV Cache Compression at Test Time
Figure 4 for Scissorhands: Exploiting the Persistence of Importance Hypothesis for LLM KV Cache Compression at Test Time
Viaarxiv icon

CARAMEL: A Succinct Read-Only Lookup Table via Compressed Static Functions

May 26, 2023
Benjamin Coleman, David Torres Ramos, Vihan Lakshman, Chen Luo, Anshumali Shrivastava

Figure 1 for CARAMEL: A Succinct Read-Only Lookup Table via Compressed Static Functions
Figure 2 for CARAMEL: A Succinct Read-Only Lookup Table via Compressed Static Functions
Figure 3 for CARAMEL: A Succinct Read-Only Lookup Table via Compressed Static Functions
Viaarxiv icon

Compress, Then Prompt: Improving Accuracy-Efficiency Trade-off of LLM Inference with Transferable Prompt

May 17, 2023
Zhaozhuo Xu, Zirui Liu, Beidi Chen, Yuxin Tang, Jue Wang, Kaixiong Zhou, Xia Hu, Anshumali Shrivastava

Figure 1 for Compress, Then Prompt: Improving Accuracy-Efficiency Trade-off of LLM Inference with Transferable Prompt
Figure 2 for Compress, Then Prompt: Improving Accuracy-Efficiency Trade-off of LLM Inference with Transferable Prompt
Figure 3 for Compress, Then Prompt: Improving Accuracy-Efficiency Trade-off of LLM Inference with Transferable Prompt
Figure 4 for Compress, Then Prompt: Improving Accuracy-Efficiency Trade-off of LLM Inference with Transferable Prompt
Viaarxiv icon

BOLT: An Automated Deep Learning Framework for Training and Deploying Large-Scale Neural Networks on Commodity CPU Hardware

Mar 30, 2023
Nicholas Meisburger, Vihan Lakshman, Benito Geordie, Joshua Engels, David Torres Ramos, Pratik Pranav, Benjamin Coleman, Benjamin Meisburger, Shubh Gupta, Yashwanth Adunukota, Tharun Medini, Anshumali Shrivastava

Figure 1 for BOLT: An Automated Deep Learning Framework for Training and Deploying Large-Scale Neural Networks on Commodity CPU Hardware
Figure 2 for BOLT: An Automated Deep Learning Framework for Training and Deploying Large-Scale Neural Networks on Commodity CPU Hardware
Figure 3 for BOLT: An Automated Deep Learning Framework for Training and Deploying Large-Scale Neural Networks on Commodity CPU Hardware
Figure 4 for BOLT: An Automated Deep Learning Framework for Training and Deploying Large-Scale Neural Networks on Commodity CPU Hardware
Viaarxiv icon

A Theoretical Analysis Of Nearest Neighbor Search On Approximate Near Neighbor Graph

Mar 10, 2023
Anshumali Shrivastava, Zhao Song, Zhaozhuo Xu

Figure 1 for A Theoretical Analysis Of Nearest Neighbor Search On Approximate Near Neighbor Graph
Figure 2 for A Theoretical Analysis Of Nearest Neighbor Search On Approximate Near Neighbor Graph
Viaarxiv icon

Learning Multimodal Data Augmentation in Feature Space

Dec 29, 2022
Zichang Liu, Zhiqiang Tang, Xingjian Shi, Aston Zhang, Mu Li, Anshumali Shrivastava, Andrew Gordon Wilson

Figure 1 for Learning Multimodal Data Augmentation in Feature Space
Figure 2 for Learning Multimodal Data Augmentation in Feature Space
Figure 3 for Learning Multimodal Data Augmentation in Feature Space
Figure 4 for Learning Multimodal Data Augmentation in Feature Space
Viaarxiv icon

The trade-offs of model size in large recommendation models : A 10000 $\times$ compressed criteo-tb DLRM model (100 GB parameters to mere 10MB)

Jul 21, 2022
Aditya Desai, Anshumali Shrivastava

Figure 1 for The trade-offs of model size in large recommendation models : A 10000 $\times$ compressed criteo-tb DLRM model (100 GB parameters to mere 10MB)
Figure 2 for The trade-offs of model size in large recommendation models : A 10000 $\times$ compressed criteo-tb DLRM model (100 GB parameters to mere 10MB)
Figure 3 for The trade-offs of model size in large recommendation models : A 10000 $\times$ compressed criteo-tb DLRM model (100 GB parameters to mere 10MB)
Figure 4 for The trade-offs of model size in large recommendation models : A 10000 $\times$ compressed criteo-tb DLRM model (100 GB parameters to mere 10MB)
Viaarxiv icon

Efficient model compression with Random Operation Access Specific Tile (ROAST) hashing

Jul 21, 2022
Aditya Desai, Keren Zhou, Anshumali Shrivastava

Figure 1 for Efficient model compression with Random Operation Access Specific Tile (ROAST) hashing
Figure 2 for Efficient model compression with Random Operation Access Specific Tile (ROAST) hashing
Figure 3 for Efficient model compression with Random Operation Access Specific Tile (ROAST) hashing
Figure 4 for Efficient model compression with Random Operation Access Specific Tile (ROAST) hashing
Viaarxiv icon

Learning to Retrieve Relevant Experiences for Motion Planning

Apr 18, 2022
Constantinos Chamzas, Aedan Cullen, Anshumali Shrivastava, Lydia E. Kavraki

Figure 1 for Learning to Retrieve Relevant Experiences for Motion Planning
Figure 2 for Learning to Retrieve Relevant Experiences for Motion Planning
Figure 3 for Learning to Retrieve Relevant Experiences for Motion Planning
Figure 4 for Learning to Retrieve Relevant Experiences for Motion Planning
Viaarxiv icon