YW
Yanzhi Wang
cs.LGcs.CVcs.AIstat.MLcs.NEcs.ARcs.CLcs.ETcs.ROcs.DC
On Valency
published · living versionsW_sm7r6rxg·v1 · currentpublished
When Does Test-Time Search Help World Action Models? An Empirical Audit of Candidate Selection and Simulator Integrity
with Jun Liu, Dajie Sun, Zhenglun Kong, Zhenjiang Fan +6
1 version
Preprints & journals
243 papers in the corpus · 2016–2026Event-Aligned Visual Action Reasoning for World Action Models2610.09427v1 · Xiaomeng Yang, Yushu Wu, Yi Gao et al.2026 · 0 citationsarXiv
CIRRA: Dual-Level Continual Instruction Reconciliation with Ongoing Execution for Embodied Robot Agents in Interactive Household Tasks2610.08862v2 · Ci Zhang, Enfu Nan, Arman Akbari et al.2026 · 0 citationsarXiv
Imagine the Future, Internalize the Gist: Efficient VLA Reasoning via Internalized Spatiotemporal Imagination2610.02626v4 · Shenglan Li, Zhendong Mi, Hengyi Zhu et al.2026 · 0 citationsarXiv
SpectralCache: Accelerating Diffusion-Based World Models via Spectral Feature Caching2610.02660v1 · Zhendong Mi, Pu Zhao, Ziyu Hu et al.2026 · 0 citationsarXiv
BitNest: Bit-Nested Speculative Decoding for Memory-Efficient LLM Inference Acceleration2610.02800v2 · Chence Yang, Ningxi Cheng, Arash Akbari et al.2026 · 0 citationsarXiv
From Patching to Pruning Visual Computation in Vision Language Models2610.03389v1 · Rahul Chowdhury, Timothy A Rupprecht, Xuan Shen et al.2026 · 0 citationsarXiv
LEAP: Learned Block-wise Evidence Retrieval for Long Audio-Video Perception2609.39938v2 · Juyi Lin, Zhiqiang Lao, Jiali Cui et al.2026 · 0 citationsarXiv
Q-WAM: 4-Bit Quantization of World Action Models with Action-Subspace Protection2609.33269v1 · Arash Akbari, Arman Akbari, Jingwu Luo et al.2026 · 0 citationsarXiv
RAGMark: A Comprehensive Framework for Benchmarking Retrieval-Augmented Generation Systems2609.05760v1 · Zlatan Feric, Amir Taherin, Bin Ren et al.2026 · 0 citationsarXiv
Dynamic Hub-and-Spoke Memory for Streaming Video Understanding2608.30294v1 · Xinru Jiang, Lin Zhao, Xi Xiao et al.2026 · 0 citationsarXiv
Albireo: Adaptive, Energy-Efficient Inference Framework for Video Object Detection on the Edge2609.29648v1 · Amir Taherin, Jos'e Cano, Bin Ren et al.2026 · 0 citationsarXiv
Hydra: Phase-Aware Workload Characterization of LLM Inference across Edge SoC Generations, Backends, and Quantization Levels2608.25053v1 · Amir Taherin, Sana Taghipour Anvari, Charles Amante et al.2026 · 0 citationsarXiv
From Retrieved Context to Runtime Control: Adaptive Compression for Edge-based RAG2608.19535v1 · Zlatan Feric, Amir Taherin, Yanzhi Wang et al.2026 · 0 citationsarXiv
Mechanistic Interpretability of Structure-Aware Numerical Reasoning in LLaMA 3.1 8B2608.18419v1 · Rahul Chowdhury, Timothy A Rupprecht, Senhao Cao et al.2026 · 0 citationsarXiv
Fine-Tuning Qwen3-27B for C-to-Rust Code Translation: A Three-Stage Curriculum of Pretraining, Debugging-Aware SFT, and Task-Specific SFT2608.13681v1 · Pu Zhao, Changdi Yang, Yixiao Chen et al.2026 · 0 citationsarXiv
SparseDitto: An Agentic Sparse Compilation Framework through Architecture-Aware Synthesis on GPUs2608.05033v3 · Shiyang Li, Guangyan Sun, Jinwei Tang et al.2026 · 0 citationsarXiv
ScAle: Attention Head Scaling as a Minimal Adapter for Spatial Reasoning in Vision Language Models2606.29579v1 · Rahul Chowdhury, Timothy A Rupprecht, Xuan Shen et al.2026 · 0 citationsarXiv
Prompt2Effect: Training-Free Image-to-Video Model Specialization via LoRA Generation2606.13971v3 · Xiaomeng Yang, Yanyu Li, Gordon Guocheng Qian et al.2026 · 0 citationsarXiv
Flash-WAM: Modality-Aware Distillation for World Action Models2606.05254v2 · Arman Akbari, Ci Zhang, Arash Akbari et al.2026 · 0 citationsarXiv
OmniMem: Scalable and Adaptive Memory Retrieval for Long Video Generation2605.30519v1 · Lin Zhao, Yushu Wu, Yifan Gong et al.2026 · 0 citationsarXiv
Palette: A Modular, Controllable, and Efficient Framework for On-demand Authorized Safety Alignment Relaxation in LLMs2605.24154v2 · Qitao Tan, Xiaoying Song, Arman Akbari et al.2026 · 0 citationsarXiv
PhyWorld: Physics-Faithful World Model for Video Generation2605.19242v1 · Pu Zhao, Juyi Lin, Timothy Rupprecht et al.2026 · 0 citationsarXiv
ActQuant: Sub-4-bit Action-Guided Quantization for Vision-Language-Action Models2605.24011v3 · Arash Akbari, Arman Akbari, Masih Eskandar et al.2026 · 0 citationsarXiv
PanoWorld: Geometry-Consistent Panoramic Video World Modeling2605.15391v1 · Le Jiang, Xiangyu Bai, Bishoy Galoaa et al.2026 · 0 citationsarXiv
PhyGround: Benchmarking Physical Reasoning in Generative World Models2605.10806v1 · Juyi Lin, Arash Akbari, Yumei He et al.2026 · 0 citationsarXiv
CUDAHercules: Benchmarking Hardware-Aware Expert-level CUDA Optimization for LLMs2605.08467v1 · Shiyang Li, Zijian Zhang, Guangyan Sun et al.2026 · 0 citationsarXiv
Human Cognition in Machines: A Unified Perspective of World Models2604.16592v2 · Timothy Rupprecht, Pu Zhao, Amir Taherin et al.2026 · 0 citationsarXiv
When Should a Robot Think? Resource-Aware Reasoning via Reinforcement Learning for Embodied Robotic Decision-Making2603.16673v5 · Jun Liu, Pu Zhao, Zhenglun Kong et al.2026 · 0 citationsarXiv
HIERAMP: Coarse-to-Fine Autoregressive Amplification for Generative Dataset Distillation2603.06932v1 · Lin Zhao, Xinru Jiang, Xi Xiao et al.2026 · 0 citationsarXiv
A Very Big Video Reasoning Suite2602.20159v4 · Maijunxian Wang, Ruisi Wang, Juyi Lin et al.2026 · 0 citationsarXiv
Effective MoE-based LLM Compression by Exploiting Heterogeneous Inter-Group Experts Routing Frequency and Information Density2602.09316v2 · Zhendong Mi, Yixiao Chen, Pu Zhao et al.2026 · 0 citationsarXiv
S2DiT: Sandwich Diffusion Transformer for Mobile Streaming Video Generation2601.12719v2 · Lin Zhao, Yushu Wu, Aleksei Lebedev et al.2026 · 0 citationsarXiv
Q-realign: Piggybacking Realignment on Quantization for Safe and Efficient LLM Deployment2601.08089v1 · Qitao Tan, Xiaoying Song, Ningxi Cheng et al.2026 · 0 citationsarXiv
From Bits to Chips: An LLM-based Hardware-Aware Quantization Agent for Streamlined Deployment of LLMs2601.03484v2 · Kaiyuan Deng, Hangyu Zheng, Minghai Qing et al.2026 · 0 citationsarXiv
Open-Source Multimodal Moxin Models with Moxin-VLM and Moxin-VLA2512.22208v2 · Pu Zhao, Arash Akbari, Xuan Shen et al.2025 · 0 citationsarXiv
Advancing time series completion via RFAMoE and MDFF2512.07873v4 · Ci Zhang, Huayu Li, Changdi Yang et al.2025 · 0 citationsarXiv
ZeroSim: Zero-Shot Analog Circuit Evaluation with Unified Transformer Embeddings2511.07658v1 · Xiaomeng Yang, Jian Gao, Yanzhi Wang et al.2025 · 1 citationarXiv
LightCache: Memory-Efficient, Training-Free Acceleration for Video Generation2510.05367v1 · Yang Xiao, Gen Li, Kaiyuan Deng et al.2025 · 0 citationsarXiv
Collaborative Compression for Large-Scale MoE Deployment on Edge2509.25689v1 · Yixiao Chen, Yanyue Xie, Ruining Yang et al.2025 · 0 citationsarXiv
CircuitSense: A Hierarchical MLLM Benchmark Bridging Visual Comprehension and Symbolic Reasoning in Engineering Design Process2509.22339v2 · Arman Akbari, Jian Gao, Yifei Zou et al.2025 · 0 citationsarXiv
Cross-Platform Scaling of Vision-Language-Action Models from Edge to Cloud GPUs2509.11480v2 · Amir Taherin, Juyi Lin, Arash Akbari et al.2025 · 3 citationsarXiv
RAGs to Riches: RAG-like Few-shot Learning for Large Language Model Role-playing2509.12168v1 · Timothy Rupprecht, Enfu Nan, Arash Akbari et al.2025 · 0 citationsarXiv
TSLA: A Task-Specific Learning Adaptation for Semantic Segmentation on Autonomous Vehicles Platform2508.12279v2 · Jun Liu, Zhenglun Kong, Pu Zhao et al.2025 · 3 citationsIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 44, no. 4, pp. 1406-1419, April 2025
Taming Diffusion Transformer for Efficient Mobile Video Generation in Seconds2507.13343v2 · Yushu Wu, Yanyu Li, Anil Kag et al.2025 · 0 citationsarXiv
VOTE: Vision-Language-Action Optimization with Trajectory Ensemble Voting2507.05116v5 · Juyi Lin, Amir Taherin, Arash Akbari et al.2025 · 0 citationsarXiv
Routing Mamba: Scaling State Space Models with Mixture-of-Experts Projection2506.18145v1 · Zheng Zhan, Liliang Ren, Shuohang Wang et al.2025 · 0 citationsarXiv
Verbal Werewolf: Engage Users with Verbalized Agentic Werewolf Game Framework2506.00160v2 · Qihui Fan, Wenbo Li, Enfu Nan et al.2025 · 0 citationsarXiv
Enabling Flexible Multi-LLM Integration for Scalable Knowledge Aggregation2505.23844v1 · Zhenglun Kong, Zheng Zhan, Shiyue Hou et al.2025 · 0 citationsarXiv
ALTER: All-in-One Layer Pruning and Temporal Expert Routing for Efficient Diffusion Generation2505.21817v1 · Xiaomeng Yang, Lei Lu, Qihui Fan et al.2025 · 1 citationarXiv
Taming Diffusion for Dataset Distillation with High Representativeness2505.18399v1 · Lin Zhao, Yushu Wu, Xinru Jiang et al.2025 · 0 citationsarXiv
Career total: 670 works. 243 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.