computational cost
The overall expenses (in terms of time and resources) associated with executing an AI algorithm, which affects the feasibility of deploying the model in real-world applications.
- A Difference-of-Convex Functions Approach to Energy-Based Iterative Reasoning
- A Diffusion Model for Regular Time Series Generation from Irregular Data with Completion and Masking
- ASDSV: Multimodal Generation Made Efficient with Approximate Speculative Diffusion and Speculative Verification
- Activity Pruning for Efficient Spiking Neural Networks
- Amortized Sampling with Transferable Normalizing Flows
- Approximating Shapley Explanations in Reinforcement Learning
- Asymmetric Duos: Sidekicks Improve Uncertainty
- Availability-aware Sensor Fusion via Unified Canonical Space
- Bilevel Optimization for Adversarial Learning Problems: Sharpness, Generation, and Beyond
- Bridging Symmetry and Robustness: On the Role of Equivariance in Enhancing Adversarial Robustness
- CoIDO: Efficient Data Selection for Visual Instruction Tuning via Coupled Importance-Diversity Optimization
- Cost-Sensitive Freeze-thaw Bayesian Optimization for Efficient Hyperparameter Tuning
- CovMatch: Cross-Covariance Guided Multimodal Dataset Distillation with Trainable Text Encoder
- Datasets, Documents, and Repetitions: The Practicalities of Unequal Data Quality
- Diffusion on Demand: Selective Caching and Modulation for Efficient Generation
- EDBench: Large-Scale Electron Density Data for Molecular Modeling
- Encoder-Decoder Diffusion Language Models for Efficient Training and Inference
- Exploring Structural Degradation in Dense Representations for Self-supervised Learning
- Fast Non-Log-Concave Sampling under Nonconvex Equality and Inequality Constraints with Landing
- FedRAM: Federated Reweighting and Aggregation for Multi-Task Learning
- Fixing It in Post: A Comparative Study of LLM Post-Training Data Quality and Model Performance
- FlashMo: Geometric Interpolants and Frequency-Aware Sparsity for Scalable Efficient Motion Generation
- From Shortcut to Induction Head: How Data Diversity Shapes Algorithm Selection in Transformers
- GSPN-2: Efficient Parallel Sequence Modeling
- Glance2Gaze: Efficient Vision-Language Models from Glance Fusion to Gaze Compression
- GraSS: Scalable Data Attribution with Gradient Sparsification and Sparse Projection
- Improving Energy Natural Gradient Descent through Woodbury, Momentum, and Randomization
- LASeR: Learning to Adaptively Select Reward Models with Multi-Arm Bandits
- Learning to Control Free-Form Soft Swimmers
- Less Greedy Equivalence Search
- Localizing Knowledge in Diffusion Transformers
- Machine Unlearning via Task Simplex Arithmetic
- MergeBench: A Benchmark for Merging Domain-Specialized LLMs
- Noise Hypernetworks: Amortizing Test-Time Compute in Diffusion Models
- OmniSVG: A Unified Scalable Vector Graphics Generation Model
- Pay Attention to Small Weights
- PocketSR: The Super-Resolution Expert in Your Pocket Mobiles
- Practical Bayes-Optimal Membership Inference Attacks
- Predicting partially observable dynamical systems via diffusion models with a multiscale inference scheme
- QSVD: Efficient Low-rank Approximation for Unified Query-Key-Value Weight Compression in Low-Precision Vision-Language Models
- QuARI: Query Adaptive Retrieval Improvement
- RAST: Reasoning Activation in LLMs via Small-model Transfer
- REN: Fast and Efficient Region Encodings from Patch-Based Image Encoders
- Redefining Experts: Interpretable Decomposition of Language Models for Toxicity Mitigation
- Rethinking Optimal Verification Granularity for Compute-Efficient Test-Time Scaling
- Retrv-R1: A Reasoning-Driven MLLM Framework for Universal and Efficient Multimodal Retrieval
- RidgeLoRA: Matrix Ridge Enhanced Low-Rank Adaptation of Large Language Models
- SHAP zero Explains Biological Sequence Models with Near-zero Marginal Cost for Future Queries
- SPACE: SPike-Aware Consistency Enhancement for Test-Time Adaptation in Spiking Neural Networks
- Sampling-Efficient Test-Time Scaling: Self-Estimating the Best-of-N Sampling in Early Decoding
- Scalable Signature Kernel Computations via Local Neumann Series Expansions
- Scaling Law with Learning Rate Annealing
- Scaling Up Active Testing to Large Language Models
- ShiQ: Bringing back Bellman to LLMs
- Structured Sparse Transition Matrices to Enable State Tracking in State-Space Models
- TADA: Improved Diffusion Sampling with Training-free Augmented DynAmics
- Think or Not? Selective Reasoning via Reinforcement Learning for Vision-Language Models
- Towards Multiscale Graph-based Protein Learning with Geometric Secondary Structural Motifs
- Unified Scaling Laws for Compressed Representations
- Universally Invariant Learning in Equivariant GNNs
- VFRTok: Variable Frame Rates Video Tokenizer with Duration-Proportional Information Assumption
- When Kernels Multiply, Clusters Unify: Fusing Embeddings with the Kronecker Product