Picture for Changsheng Zhao

Changsheng Zhao

ParetoQ: Scaling Laws in Extremely Low-bit LLM Quantization

Add code
Feb 04, 2025
Viaarxiv icon

Llama Guard 3-1B-INT4: Compact and Efficient Safeguard for Human-AI Conversations

Add code
Nov 18, 2024
Viaarxiv icon

LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding

Add code
Oct 22, 2024
Figure 1 for LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding
Figure 2 for LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding
Figure 3 for LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding
Figure 4 for LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding
Viaarxiv icon

Agent-as-a-Judge: Evaluate Agents with Agents

Add code
Oct 14, 2024
Figure 1 for Agent-as-a-Judge: Evaluate Agents with Agents
Figure 2 for Agent-as-a-Judge: Evaluate Agents with Agents
Figure 3 for Agent-as-a-Judge: Evaluate Agents with Agents
Figure 4 for Agent-as-a-Judge: Evaluate Agents with Agents
Viaarxiv icon

Scaling Parameter-Constrained Language Models with Quality Data

Add code
Oct 04, 2024
Figure 1 for Scaling Parameter-Constrained Language Models with Quality Data
Figure 2 for Scaling Parameter-Constrained Language Models with Quality Data
Figure 3 for Scaling Parameter-Constrained Language Models with Quality Data
Figure 4 for Scaling Parameter-Constrained Language Models with Quality Data
Viaarxiv icon

BUPTCMCC-6G-CMG+: A GBSM-Based ISAC Channel Model Simulator

Add code
Sep 22, 2024
Figure 1 for BUPTCMCC-6G-CMG+: A GBSM-Based ISAC Channel Model Simulator
Figure 2 for BUPTCMCC-6G-CMG+: A GBSM-Based ISAC Channel Model Simulator
Figure 3 for BUPTCMCC-6G-CMG+: A GBSM-Based ISAC Channel Model Simulator
Figure 4 for BUPTCMCC-6G-CMG+: A GBSM-Based ISAC Channel Model Simulator
Viaarxiv icon

SpinQuant: LLM quantization with learned rotations

Add code
May 28, 2024
Figure 1 for SpinQuant: LLM quantization with learned rotations
Figure 2 for SpinQuant: LLM quantization with learned rotations
Figure 3 for SpinQuant: LLM quantization with learned rotations
Figure 4 for SpinQuant: LLM quantization with learned rotations
Viaarxiv icon

Basis Selection: Low-Rank Decomposition of Pretrained Large Language Models for Target Applications

Add code
May 24, 2024
Viaarxiv icon

MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases

Add code
Feb 22, 2024
Viaarxiv icon

Not All Weights Are Created Equal: Enhancing Energy Efficiency in On-Device Streaming Speech Recognition

Add code
Feb 20, 2024
Figure 1 for Not All Weights Are Created Equal: Enhancing Energy Efficiency in On-Device Streaming Speech Recognition
Figure 2 for Not All Weights Are Created Equal: Enhancing Energy Efficiency in On-Device Streaming Speech Recognition
Figure 3 for Not All Weights Are Created Equal: Enhancing Energy Efficiency in On-Device Streaming Speech Recognition
Figure 4 for Not All Weights Are Created Equal: Enhancing Energy Efficiency in On-Device Streaming Speech Recognition
Viaarxiv icon