Optimization

Temporal Scaling Law for Large Language Models EMNLP 2025 TermDiffuSum: A Term-guided Diffusion Model for Extractive Summarization of Legal Documents COLING 2025 Test-Time Adaptation on Noisy Data via Model-Pruning-Based Filtering and Flatness-Aware Entropy Minimization AAAI 2025 Test-time Alignment of Diffusion Models without Reward Over-optimization ICLR 2025 Test-Time Canonicalization by Foundation Models for Robust Perception ICML 2025 Test-Time Steering for Lossless Text Compression via Weighted Product of Experts EMNLP 2025 Test-Time Training Provably Improves Transformers as In-context Learners ICML 2025 Text2PDE: Latent Diffusion Models for Accessible Physics Simulation ICLR 2025 TGDPO: Harnessing Token-Level Reward Guidance for Enhancing Direct Preference Optimization ICML 2025 The Adaptive Complexity of Finding a Stationary Point COLT 2025 The Adaptive Q-Network for Recommendation Tasks with Dynamic Item Space AAAI 2025 The AdEMAMix Optimizer: Better, Faster, Older ICLR 2025 The Batch Complexity of Bandit Pure Exploration ICML 2025 The Best of Both Worlds: Bridging Quality and Diversity in Data Selection with Bipartite Graph ICML 2025 The Butterfly Effect: Neural Network Training Trajectories Are Highly Sensitive to Initial Conditions ICML 2025 The Complexity of Extending Fair Allocations of Indivisible Goods AAAI 2025 The Crucial Role of Samplers in Online Direct Preference Optimization ICLR 2025 The Curse of Conditions: Analyzing and Improving Optimal Transport for Conditional Flow-Based Generation ICCV 2025 The Devil Is in the Details: Tackling Unimodal Spurious Correlations for Generalizable Multimodal Reward Models ICML 2025 The First Theoretical Approximation Guarantees for the Non-Dominated Sorting Genetic Algorithm III (NSGA-III) IJCAI 2025 The Global Convergence Time of Stochastic Gradient Descent in Non-Convex Landscapes: Sharp Estimates via Large Deviations ICML 2025 The Gradient of Algebraic Model Counting AAAI 2025 The Impact of On-Policy Parallelized Data Collection on Deep Reinforcement Learning Networks ICML 2025 The Journey Matters: Average Parameter Count over Pre-training Unifies Sparse and Dense Scaling Laws ICLR 2025 The late-stage training dynamics of (stochastic) subgradient descent on homogeneous neural networks COLT 2025