optimization
The mathematical and algorithmic approaches used to improve the performance of AI models, involving the selection of optimal parameters to minimize or maximize objective functions.
- 3BASiL: An Algorithmic Framework for Sparse plus Low-Rank Compression of LLMs
- A Tale of Two Symmetries: Exploring the Loss Landscape of Equivariant Models
- AdvPrefix: An Objective for Nuanced LLM Jailbreaks
- Balancing Multimodal Training Through Game-Theoretic Regularization
- Beyond Prediction: Managing the Repercussions of Machine Learning Applications
- Boosting Adversarial Transferability with Spatial Adversarial Alignment
- Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs
- Deep Legendre Transform
- Delving into Cascaded Instability: A Lipschitz Continuity View on Image Restoration and Object Detection Synergy
- Does Representation Guarantee Welfare?
- Drag-and-Drop LLMs: Zero-Shot Prompt-to-Weights
- Dynamic Bundling with Large Language Models for Zero-Shot Inference on Text-Attributed Graphs
- Effective Policy Learning for Multi-Agent Online Coordination Beyond Submodular Objectives
- Embedding Principle of Homogeneous Neural Network for Classification Problem
- Fit the Distribution: Cross-Image/Prompt Adversarial Attacks on Multimodal Large Language Models
- Fix False Transparency by Noise Guided Splatting
- GUI-G1: Understanding R1-Zero-Like Training for Visual Grounding in GUI Agents
- Generalized Linear Mode Connectivity for Transformers
- GradMetaNet: An Equivariant Architecture for Learning on Gradients
- Gradient Variance Reveals Failure Modes in Flow-Based Generative Models
- Hamiltonian Descent Algorithms for Optimization: Accelerated Rates via Randomized Integration Time
- Handling Label Noise via Instance-Level Difficulty Modeling and Dynamic Optimization
- IMPACT: Irregular Multi-Patch Adversarial Composition Based on Two‑Phase Optimization
- Impact of Layer Norm on Memorization and Generalization in Transformers
- Improving Model-Based Reinforcement Learning by Converging to Flatter Minima
- In Search of Adam’s Secret Sauce
- In Search of Adam’s Secret Sauce
- Instant4D: 4D Gaussian Splatting in Minutes
- LARGO: Latent Adversarial Reflection through Gradient Optimization for Jailbreaking LLMs
- Learning Provably Improves the Convergence of Gradient Descent
- Learning quadratic neural networks in high dimensions: SGD dynamics and scaling laws
- Look-Ahead Reasoning on Learning Platforms
- MGUP: A Momentum-Gradient Alignment Update Policy for Stochastic Optimization
- MURKA: Multi-Reward Reinforcement Learning with Knowledge Alignment for Optimization Tasks
- Noise Hypernetworks: Amortizing Test-Time Compute in Diffusion Models
- Online Two-Stage Submodular Maximization
- PAC-Bayes Bounds for Multivariate Linear Regression and Linear Autoencoders
- Pass@K Policy Optimization: Solving Harder Reinforcement Learning Problems
- Query-Efficient Locally Private Hypothesis Selection via the Scheffe Graph
- RF-Agent: Automated Reward Function Design via Language Agent Tree Search
- Rethinking Fair Federated Learning from Parameter and Client View
- Rethinking Multimodal Learning from the Perspective of Mitigating Classification Ability Disproportion
- Reward-oriented Causal Representation Learning
- RobotSmith: Generative Robotic Tool Design for Acquisition of Complex Manipulation Skills
- Robust Satisficing Gaussian Process Bandits Under Adversarial Attacks
- Rooms from Motion: Un-posed Indoor 3D Object Detection as Localization and Mapping
- Sharp Gaussian approximations for Decentralized Federated Learning
- Test3R: Learning to Reconstruct 3D at Test Time
- Thompson Sampling for Multi-Objective Linear Contextual Bandit
- Towards Generalizable 3D Human Pose Estimation via Ensembles on Flat Loss Landscapes
- Tree of Preferences for Diversified Recommendation
- TreeSplat: Mergeable Tree for Deformable Gaussian Splatting