Optimization

FlatQuant: Flatness Matters for LLM Quantization ICML 2025 Flat Reward in Policy Parameter Space Implies Robust Reinforcement Learning ICLR 2025 Flexible and Efficient Grammar-Constrained Decoding ICML 2025 Flexible-length Text Infilling for Discrete Diffusion Models EMNLP 2025 Flexible Sharpness-Aware Personalized Federated Learning AAAI 2025 FlexiGPT: Pruning and Extending Large Language Models with Low-Rank Weight Sharing NAACL 2025 FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference ICLR 2025 Flow Distillation Sampling: Regularizing 3D Gaussians with Pre-trained Matching Priors ICLR 2025 FlowDPS : Flow-Driven Posterior Sampling for Inverse Problems ICCV 2025 Flowing Datasets with Wasserstein over Wasserstein Gradient Flows ICML 2025 FlowMamba: Learning Point Cloud Scene Flow with Global Motion Propagation AAAI 2025 Flow matching achieves almost minimax optimal convergence ICLR 2025 Flow Matching for Denoised Social Recommendation ICML 2025 Floxels: Fast Unsupervised Voxel Based Scene Flow Estimation CVPR 2025 FLRC: Fine-grained Low-Rank Compressor for Efficient LLM Inference EMNLP 2025 FNIN: A Fourier Neural Operator-based Numerical Integration Network for Surface-from-gradients AAAI 2025 FocalPO: Enhancing Preference Optimizing by Focusing on Correct Preference Rankings ACL 2025 Focal-SAM: Focal Sharpness-Aware Minimization for Long-Tailed Classification ICML 2025 Focused-DPO: Enhancing Code Generation Through Focused Preference Optimization on Error-Prone Points ACL 2025 Forewarned is Forearmed: Pre-Synthesizing Jailbreak-like Instructions to Enhance LLM Safety Guardrail to Potential Attacks EMNLP 2025 FORGETTER with forgetful hyperparameters and recurring sleeps can continue to learn beyond normal overtfitting limits EMNLP 2025 Formalizing Test-Time Compute for Function-Level Code Generation IJCNLP 2025 FPE2M2: Approaching Lossless and Efficient Quantization with Native Floating Point ACL 2025 FRAME: Boosting LLMs with A Four-Quadrant Multi-Stage Pretraining Strategy ACL 2025 FreCaS: Efficient Higher-Resolution Image Generation via Frequency-aware Cascaded Sampling ICLR 2025