Zechun Liu

MobileLLM-Flash: Latency-Guided On-Device LLM Design for Industry Scale Deployment ACL 2026 VideoAuto-R1: Video Auto Reasoning via Thinking Once, Answering Twice CVPR 2026 Agent-as-a-Judge: Evaluate Agents with Agents ICML 2025 Efficient Track Anything ICCV 2025 LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding ICML 2025 ParamΔ for Direct Mixing: Post-Train Large Language Model At Zero Cost ICLR 2025 PARQ: Piecewise-Affine Regularized Quantization ICML 2025 R-Sparse: Rank-Aware Activation Sparsity for Efficient LLM Inference ICLR 2025 SpinQuant: LLM Quantization with Learned Rotations ICLR 2025 LLM-QAT: Data-Free Quantization Aware Training for Large Language Models ACL 2024 Mixture-of-Supernets: Improving Weight-Sharing Supernet Training with Architecture-Routed Mixture-of-Experts ACL 2024 MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases ICML 2024 RoLoRA: Fine-tuning Rotated Outlier-free LLMs for Effective Weight-Activation Quantization EMNLP 2024 Scaling Parameter-Constrained Language Models with Quality Data EMNLP 2024 Target-Aware Language Modeling via Granular Data Sampling EMNLP 2024 Binary and Ternary Natural Language Generation ACL 2023 LLM-FP4: 4-Bit Floating-Point Quantized Transformers EMNLP 2023 Oscillation-free Quantization for Low-bit Vision Transformers ICML 2023 BiT: Robustly Binarized Multi-distilled Transformer NeurIPS 2022 Data-Free Neural Architecture Search via Recursive Label Calibration ECCV 2022 Nonuniform-to-Uniform Quantization: Towards Accurate Quantization via Generalized Straight-Through Estimation CVPR 2022 SDQ: Stochastic Differentiable Quantization with Mixed Precision ICML 2022 Sliced Recursive Transformer ECCV 2022 Stereo Neural Vernier Caliper AAAI 2022 Un-mix: Rethinking Image Mixtures for Unsupervised Visual Representation Learning AAAI 2022 Vision Transformer Slimming: Multi-Dimension Searching in Continuous Optimization Space CVPR 2022 Conditional Link Prediction of Category-Implicit Keypoint Detection WACV 2021 How Do Adam and Training Strategies Help BNNs Optimization ICML 2021 Is Label Smoothing Truly Incompatible with Knowledge Distillation: An Empirical Study ICLR 2021 Partial Is Better Than All: Revisiting Fine-tuning Strategy for Few-shot Learning AAAI 2021 S2-BNN: Bridging the Gap Between Self-Supervised Real and 1-Bit Neural Networks via Guided Distribution Calibration CVPR 2021 Binarizing MobileNet via Evolution-Based Searching CVPR 2020 ReActNet: Towards Precise Binary Neural Network with Generalized Activation Functions ECCV 2020 Single Path One-Shot Neural Architecture Search with Uniform Sampling ECCV 2020 Latent Weights Do Not Exist: Rethinking Binarized Neural Network Optimization NeurIPS 2019 MetaPruning: Meta Learning for Automatic Neural Network Channel Pruning ICCV 2019 Bi-Real Net: Enhancing the Performance of 1-bit CNNs with Improved Representational Capability and Advanced Training Algorithm ECCV 2018