Skip To Main Content
Chun Jason on Muck Rack

Chun Jason

As seen in: arXiv
Is this you? As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work. Claim your profile
Back to Chun Jason’s profile

Articles by Chun Jason

Computer Science > Computation and Language

|
By Chun Jason
| arXiv
Speculative decoding accelerates large language model inference through collaboration between a lightweight draft model and a target verifier. Existing methods mainly improve the draft side, while the target model is typically kept dense and unchanged. We show that, under domain-specific inference, full-depth target verification is not always the optimal choice.
Open in Who Shared Issue with article?

Computer Science > Distributed, Parallel, and Cluster Computing

|
By Chun Jason
| arXiv
Vision-Language-Action (VLA) models are becoming a promising paradigm for autonomous driving, but their deployment on existing vehicle platforms remains difficult because they introduce both high inference latency and strong GPU-side resource pressure.
Open in Who Shared Issue with article?

Computer Science > Computation and Language

|
By Chun Jason
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (638) Issue with article?

ClawMobile: Rethinking Smartphone-Native Agentic Systems

|
By Chun Jason
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (282) Issue with article?

ReFilter: Improving Robustness of Retrieval-Augmented Generation via Gated Filter

|
By Chun Jason
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (421) Issue with article?

[2511.19986] On-Demand Multi-Task Sparsity for Efficient Large-Model Deployment on Edge Devices

|
By Chun Jason, Haibo Hu, Qiao Li, Nan Guan
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (513) Issue with article?

DeeAD: Dynamic Early Exit of Vision-Language Action for Efficient Autonomous Driving

|
By Chun Jason
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (665) Issue with article?

Nav-EE: Navigation-Guided Early Exiting for Efficient Vision-Language Models in Autonomous Driving

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

Beyond Semantic Similarity: Reducing Unnecessary API Calls via Behavior-Aligned Retriever

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

[2507.09642] Rethinking LSM-tree based Key-Value Stores: A Survey

|
By Chun Jason, Qiao Li, Quanqing Xu, Congming Gao
| arXiv
Open in Who Shared Issue with article?

AD-EE: Early Exiting for Fast and Reliable Vision-Language Models in Autonomous Driving

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

[2505.06297] Lossless Compression of Large Language Model-Generated Text via Next-Token Prediction

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

Easz: An Agile Transformer-based Image Compression Framework for Resource-constrained IoTs

|
By Chun Jason
| arXiv
Work on one of the world's most important websites and make an impact on open science.
Open in Who Shared View Duplicates (548) Issue with article?

VLM-C4L: Continual Core Dataset Learning with Corner Case Optimization via Vision-Language Models for Autonomous Driving

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

FlexInfer: Breaking Memory Constraint via Flexible and Efficient Offloading for On-Device LLM Inference

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

CoT-VLM4Tar: Chain-of-Thought Guided Vision-Language Models for Traffic Anomaly Resolution

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

When Compression Meets Model Compression: Memory-Efficient Double Compression for Large Language Models

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

EvoP: Robust LLM Inference via Evolutionary Pruning

|
By Chun Jason
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (431) Issue with article?

A$^2$ATS: Retrieval-Based KV Cache Reduction via Windowed Rotary Position Embedding and Query-Aware Vector Quantization

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

Ariadne: A Hotness-Aware and Size-Adaptive Compressed Swap Technique for Fast Application Relaunch and Reduced CPU Usage on Mobile Devices

|
By Chun Jason, Nika Mansouri
| arXiv
Open in Who Shared Issue with article?

RALAD: Bridging the Real-to-Sim Domain Gap in Autonomous Driving with Retrieval-Augmented Learning

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

GeneQuery: A General QA-based Framework for Spatial Gene Expression Predictions from Histology Images

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

SHAP-CAT: A interpretable multi-modal framework enhancing WSI classification via virtual staining and shapley-value-based multimodal fusion

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

Advances in Multiple Instance Learning for Whole Slide Image Analysis: Techniques, Challenges, and Future Directions

|
By Chun Jason
| arXiv
Coming this September, the Forum is free, virtual, and open to all. Sign Up and Learn more.
Open in Who Shared View Duplicates (610) Issue with article?

[2407.13193] Retrieval-Augmented Generation for Natural Language Processing: A Survey

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

Grade Like a Human: Rethinking Automated Assessment with Large Language Models

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

BehaviorGPT: Smart Agent Simulation for Autonomous Driving with Next-Patch Prediction

|
By Chun Jason, Haibo Hu, Xinhong Chen, Jianping Wang
| arXiv
[Submitted on 27 May 2024 ( v1 ), last revised 8 Nov 2024 (this version, v2)] Title:BehaviorGPT: Smart Agent Simulation for Autonomous Driving with Next-Patch Prediction View a PDF of the paper titled BehaviorGPT: Smart Agent Simulation for Autonomous Driving with Next-Patch Prediction, by Zikang Zhou and 8 other authors View PDF HTML (experimental) Abstract:Simulating realistic interactions among traffic agents is crucial for efficiently validating the safety of autonomous driving systems.
Open in Who Shared Issue with article?

RAEE: A Robust Retrieval-Augmented Early Exiting Framework for Efficient Inference

|
By Chun Jason
| arXiv
[Submitted on 24 May 2024 ( v1 ), last revised 20 Sep 2024 (this version, v2)] Title:RAEE: A Robust Retrieval-Augmented Early Exiting Framework for Efficient Inference View a PDF of the paper titled RAEE: A Robust Retrieval-Augmented Early Exiting Framework for Efficient Inference, by Lianming Huang and 7 other authors View PDF HTML (experimental) Abstract:Deploying large language model inference remains challenging due to their high computational overhead.
Open in Who Shared Issue with article?

Pre-processing matters: A segment search method for WSI classification

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

[2401.02993] Improving Natural Language Understanding with Computation-Efficient Retrieval Representation Fusion

|
By Chun Jason
| arXiv
Open in Who Shared Issue with article?

Moby: Empowering 2D Models for Efficient Point Cloud Analytics on the Edge

|
By Yik Hong, Chun Jason
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (591) Issue with article?
Show More