Picture for Beidi Chen

Beidi Chen

On the Surprising Effectiveness of Attention Transfer for Vision Transformers

Add code
Nov 14, 2024
Viaarxiv icon

ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference

Add code
Oct 28, 2024
Viaarxiv icon

MagicPIG: LSH Sampling for Efficient LLM Generation

Add code
Oct 21, 2024
Figure 1 for MagicPIG: LSH Sampling for Efficient LLM Generation
Figure 2 for MagicPIG: LSH Sampling for Efficient LLM Generation
Figure 3 for MagicPIG: LSH Sampling for Efficient LLM Generation
Figure 4 for MagicPIG: LSH Sampling for Efficient LLM Generation
Viaarxiv icon

Model-GLUE: Democratized LLM Scaling for A Large Model Zoo in the Wild

Add code
Oct 07, 2024
Figure 1 for Model-GLUE: Democratized LLM Scaling for A Large Model Zoo in the Wild
Figure 2 for Model-GLUE: Democratized LLM Scaling for A Large Model Zoo in the Wild
Figure 3 for Model-GLUE: Democratized LLM Scaling for A Large Model Zoo in the Wild
Figure 4 for Model-GLUE: Democratized LLM Scaling for A Large Model Zoo in the Wild
Viaarxiv icon

Sirius: Contextual Sparsity with Correction for Efficient LLMs

Add code
Sep 05, 2024
Viaarxiv icon

MagicDec: Breaking the Latency-Throughput Tradeoff for Long Context Generation with Speculative Decoding

Add code
Aug 21, 2024
Figure 1 for MagicDec: Breaking the Latency-Throughput Tradeoff for Long Context Generation with Speculative Decoding
Figure 2 for MagicDec: Breaking the Latency-Throughput Tradeoff for Long Context Generation with Speculative Decoding
Figure 3 for MagicDec: Breaking the Latency-Throughput Tradeoff for Long Context Generation with Speculative Decoding
Figure 4 for MagicDec: Breaking the Latency-Throughput Tradeoff for Long Context Generation with Speculative Decoding
Viaarxiv icon

MINI-SEQUENCE TRANSFORMER: Optimizing Intermediate Memory for Long Sequences Training

Add code
Jul 22, 2024
Figure 1 for MINI-SEQUENCE TRANSFORMER: Optimizing Intermediate Memory for Long Sequences Training
Figure 2 for MINI-SEQUENCE TRANSFORMER: Optimizing Intermediate Memory for Long Sequences Training
Figure 3 for MINI-SEQUENCE TRANSFORMER: Optimizing Intermediate Memory for Long Sequences Training
Figure 4 for MINI-SEQUENCE TRANSFORMER: Optimizing Intermediate Memory for Long Sequences Training
Viaarxiv icon

VcLLM: Video Codecs are Secretly Tensor Codecs

Add code
Jun 29, 2024
Figure 1 for VcLLM: Video Codecs are Secretly Tensor Codecs
Figure 2 for VcLLM: Video Codecs are Secretly Tensor Codecs
Figure 3 for VcLLM: Video Codecs are Secretly Tensor Codecs
Figure 4 for VcLLM: Video Codecs are Secretly Tensor Codecs
Viaarxiv icon

It Takes Two: On the Seamlessness between Reward and Policy Model in RLHF

Add code
Jun 12, 2024
Viaarxiv icon

LoCoCo: Dropping In Convolutions for Long Context Compression

Add code
Jun 08, 2024
Viaarxiv icon