SOTAVerified

Computational Efficiency

Methods and optimizations to reduce the computational resources (e.g., time, memory, or power) needed for training and inference in models. This involves techniques that streamline processing, optimize algorithms, or leverage hardware to enhance performance without compromising accuracy.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 151–175 of 4891 papers

TitleStatusHype
HeadInfer: Memory-Efficient LLM Inference by Head-wise OffloadingCode2
Hybrid 3D-4D Gaussian Splatting for Fast Dynamic Scene RepresentationCode2
AdaFisher: Adaptive Second Order Optimization via Fisher InformationCode2
Grappa -- A Machine Learned Molecular Mechanics Force FieldCode2
BiFormer: Vision Transformer with Bi-Level Routing AttentionCode2
Harder Tasks Need More Experts: Dynamic Routing in MoE ModelsCode2
A Light-Weight Framework for Open-Set Object Detection with Decoupled Feature Alignment in Joint SpaceCode2
BitDecoding: Unlocking Tensor Cores for Long-Context LLMs Decoding with Low-Bit KV CacheCode2
GoMAvatar: Efficient Animatable Human Modeling from Monocular Video Using Gaussians-on-MeshCode2
Agent Attention: On the Integration of Softmax and Linear AttentionCode2
L4acados: Learning-based models for acados, applied to Gaussian process-based predictive controlCode2
LandMarkSystem Technical ReportCode2
Latent Neural Operator for Solving Forward and Inverse PDE ProblemsCode2
L-AutoDA: Leveraging Large Language Models for Automated Decision-based Adversarial AttacksCode2
Accelerating Direct Preference Optimization with Prefix SharingCode2
ClearSight: Visual Signal Enhancement for Object Hallucination Mitigation in Multimodal Large language ModelsCode2
GotenNet: Rethinking Efficient 3D Equivariant Graph Neural NetworksCode2
Generalized and Efficient 2D Gaussian Splatting for Arbitrary-scale Super-ResolutionCode2
FuXi Weather: A data-to-forecast machine learning system for global weatherCode2
Geometry Aware Operator Transformer as an Efficient and Accurate Neural Surrogate for PDEs on Arbitrary DomainsCode2
Flow Matching in Latent SpaceCode2
LoRA-IR: Taming Low-Rank Experts for Efficient All-in-One Image RestorationCode2
BEBLID: Boosted efficient binary local image descriptorCode2
FLowHigh: Towards Efficient and High-Quality Audio Super-Resolution with Single-Step Flow MatchingCode2
Free Video-LLM: Prompt-guided Visual Perception for Efficient Training-free Video LLMsCode2
Show:102550
← PrevPage 7 of 196Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ViTaLHamming Loss0.05—Unverified