Optimization

A Survey on Efficient Large Language Model Training: From Data-centric Perspectives ACL 2025 Asymmetric Learning for Spectral Graph Neural Networks AAAI 2025 Asymptotic Unbiased Sample Sampling to Speed Up Sharpness-Aware Minimization AAAI 2025 AtomNet: Designing Tiny Models from Operators Under Extreme MCU Constraints AAAI 2025 ATP: Adaptive Threshold Pruning for Efficient Data Encoding in Quantum Neural Networks CVPR 2025 Attack-inspired Calibration Loss for Calibrating Crack Recognition AAAI 2025 Augmenting Perceptual Super-Resolution via Image Quality Predictors CVPR 2025 AutoBreach: Universal and Adaptive Jailbreaking with Efficient Wordplay-Guided Optimization via Multi-LLMs NAACL 2025 Automated Fine-Grained Mixture-of-Experts Quantization ACL 2025 AutoMixer: Checkpoint Artifacts as Automatic Data Mixers ACL 2025 Balanced Rate-Distortion Optimization in Learned Image Compression CVPR 2025 Balanced Sharpness-Aware Minimization for Imbalanced Regression ICCV 2025 Balls-and-Bins Sampling for DP-SGD AISTATS 2025 Better Embeddings with Coupled Adam ACL 2025 Beyond Fixed Length: Bucket Pre-training is All You Need IJCAI 2025 BIGE : Biomechanics-informed GenAI for Exercise Science L4DC 2025 Black-box Optimization with Unknown Constraints via Overparameterized Deep Neural Networks UAI 2025 BlockDance: Reuse Structurally Similar Spatio-Temporal Features to Accelerate Diffusion Transformers CVPR 2025 Boost Your Human Image Generation Model via Direct Preference Optimization CVPR 2025 Both Asymptotic and Non-Asymptotic Convergence of Quasi-Hyperbolic Momentum using Increasing Batch Size ACML 2025 Bypass Back-propagation: Optimization-based Structural Pruning for Large Language Models via Policy Gradient ACL 2025 CalibraEval: Calibrating Prediction Distribution to Mitigate Selection Bias in LLMs-as-Judges ACL 2025 Calibration Across Layers: Understanding Calibration Evolution in LLMs EMNLP 2025 CLaSp: In-Context Layer Skip for Self-Speculative Decoding ACL 2025 CoMMIT: Coordinated Multimodal Instruction Tuning EMNLP 2025