Picture for Yifan Qiao

Yifan Qiao

ConServe: Harvesting GPUs for Low-Latency and High-Throughput Large Language Model Serving

Add code
Oct 02, 2024
Viaarxiv icon

Weighted KL-Divergence for Document Ranking Model Refinement

Add code
Jun 10, 2024
Viaarxiv icon

Approximate Cluster-Based Sparse Document Retrieval with Segmented Maximum Term Weights

Add code
Apr 13, 2024
Viaarxiv icon

Representation Sparsification with Hybrid Thresholding for Fast SPLADE-based Document Retrieval

Add code
Jun 20, 2023
Figure 1 for Representation Sparsification with Hybrid Thresholding for Fast SPLADE-based Document Retrieval
Figure 2 for Representation Sparsification with Hybrid Thresholding for Fast SPLADE-based Document Retrieval
Figure 3 for Representation Sparsification with Hybrid Thresholding for Fast SPLADE-based Document Retrieval
Figure 4 for Representation Sparsification with Hybrid Thresholding for Fast SPLADE-based Document Retrieval
Viaarxiv icon

Optimizing Guided Traversal for Fast Learned Sparse Retrieval

Add code
May 02, 2023
Figure 1 for Optimizing Guided Traversal for Fast Learned Sparse Retrieval
Figure 2 for Optimizing Guided Traversal for Fast Learned Sparse Retrieval
Figure 3 for Optimizing Guided Traversal for Fast Learned Sparse Retrieval
Figure 4 for Optimizing Guided Traversal for Fast Learned Sparse Retrieval
Viaarxiv icon

Bamboo: Making Preemptible Instances Resilient for Affordable Training of Large DNNs

Add code
Apr 26, 2022
Figure 1 for Bamboo: Making Preemptible Instances Resilient for Affordable Training of Large DNNs
Figure 2 for Bamboo: Making Preemptible Instances Resilient for Affordable Training of Large DNNs
Figure 3 for Bamboo: Making Preemptible Instances Resilient for Affordable Training of Large DNNs
Figure 4 for Bamboo: Making Preemptible Instances Resilient for Affordable Training of Large DNNs
Viaarxiv icon

Dual Skipping Guidance for Document Retrieval with Learned Sparse Representations

Add code
Apr 23, 2022
Figure 1 for Dual Skipping Guidance for Document Retrieval with Learned Sparse Representations
Figure 2 for Dual Skipping Guidance for Document Retrieval with Learned Sparse Representations
Figure 3 for Dual Skipping Guidance for Document Retrieval with Learned Sparse Representations
Figure 4 for Dual Skipping Guidance for Document Retrieval with Learned Sparse Representations
Viaarxiv icon

Compact Token Representations with Contextual Quantization for Efficient Document Re-ranking

Add code
Mar 29, 2022
Figure 1 for Compact Token Representations with Contextual Quantization for Efficient Document Re-ranking
Figure 2 for Compact Token Representations with Contextual Quantization for Efficient Document Re-ranking
Figure 3 for Compact Token Representations with Contextual Quantization for Efficient Document Re-ranking
Figure 4 for Compact Token Representations with Contextual Quantization for Efficient Document Re-ranking
Viaarxiv icon

Dorylus: Affordable, Scalable, and Accurate GNN Training with Distributed CPU Servers and Serverless Threads

Add code
May 25, 2021
Figure 1 for Dorylus: Affordable, Scalable, and Accurate GNN Training with Distributed CPU Servers and Serverless Threads
Figure 2 for Dorylus: Affordable, Scalable, and Accurate GNN Training with Distributed CPU Servers and Serverless Threads
Figure 3 for Dorylus: Affordable, Scalable, and Accurate GNN Training with Distributed CPU Servers and Serverless Threads
Figure 4 for Dorylus: Affordable, Scalable, and Accurate GNN Training with Distributed CPU Servers and Serverless Threads
Viaarxiv icon

Composite Re-Ranking for Efficient Document Search with BERT

Add code
Mar 12, 2021
Figure 1 for Composite Re-Ranking for Efficient Document Search with BERT
Figure 2 for Composite Re-Ranking for Efficient Document Search with BERT
Figure 3 for Composite Re-Ranking for Efficient Document Search with BERT
Figure 4 for Composite Re-Ranking for Efficient Document Search with BERT
Viaarxiv icon