Image de couverture de livre

RanDeS: Randomized Delta Superposition for Multi-Model Compression

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Zhou, Hangyu, Gokaslan, Aaron, Kuleshov, Volodymyr, Hariharan, Bharath
Format:	Preprint
Publié:	2025
Sujets:	Machine Learning Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Encoder-Decoder Diffusion Language Models for Efficient Training and Inference
par: Arriola, Marianne, et autres
Publié: (2025)

The Diffusion Duality
par: Sahoo, Subham Sekhar, et autres
Publié: (2025)

Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models
par: Arriola, Marianne, et autres
Publié: (2025)

Simple and Effective Masked Diffusion Language Models
par: Sahoo, Subham Sekhar, et autres
Publié: (2024)

The GAN is dead; long live the GAN! A Modern GAN Baseline
par: Huang, Yiwen, et autres
Publié: (2025)

Active Preference Inference using Language Models and Probabilistic Reasoning
par: Piriyakulkij, Wasu Top, et autres
Publié: (2023)

Counter-Current Learning: A Biologically Plausible Dual Network Approach for Deep Learning
par: Kao, Chia-Hsiang, et autres
Publié: (2024)

Diffusion Models With Learned Adaptive Noise
par: Sahoo, Subham Sekhar, et autres
Publié: (2023)

Superposition as Lossy Compression: Measure with Sparse Autoencoders and Connect to Adversarial Vulnerability
par: Bereska, Leonard, et autres
Publié: (2025)

ModuLoRA: Finetuning 2-Bit LLMs on Consumer GPUs by Integrating with Modular Quantizers
par: Yin, Junjie, et autres
Publié: (2023)

Mathematical Models of Computation in Superposition
par: Hänni, Kaarel, et autres
Publié: (2024)

Caduceus: Bi-Directional Equivariant Long-Range DNA Sequence Modeling
par: Schiff, Yair, et autres
Publié: (2024)

DeltaLLM: Compress LLMs with Low-Rank Deltas between Shared Weights
par: Mikaelyan, Liana, et autres
Publié: (2025)

QuIP#: Even Better LLM Quantization with Hadamard Incoherence and Lattice Codebooks
par: Tseng, Albert, et autres
Publié: (2024)

Parameter-Efficient Neural CDEs via Implicit Function Jacobians
par: Kuleshov, Ilya, et autres
Publié: (2025)

Superposition in Graph Neural Networks
par: Pertl, Lukas, et autres
Publié: (2025)

Sparsity and Superposition in Mixture of Experts
par: Chaudhari, Marmik, et autres
Publié: (2025)

DeNOTS: Stable Deep Neural ODEs for Time Series
par: Kuleshov, Ilya, et autres
Publié: (2024)

SineLoRA$Δ$: Sine-Activated Delta Compression
par: Gordon, Cameron, et autres
Publié: (2025)

On Implications of Scaling Laws on Feature Superposition
par: Katta, Pavan
Publié: (2024)

Towards Family-Grouped Hierarchical Federated Learning on Sub-5KB Models: A Feasibility Study of Privacy-Preserving ECG Monitoring for Ultra-Resource-Constrained Wearables
par: Wu, Hangyu
Publié: (2026)

The Geometry of Multi-Task Grokking: Transverse Instability, Superposition, and Weight Decay Phase Structure
par: Xu, Yongzhong
Publié: (2026)

DeltaDQ: Ultra-High Delta Compression for Fine-Tuned LLMs via Group-wise Dropout and Separate Quantization
par: Jiang, Yanfeng, et autres
Publié: (2024)

DAQ: Delta-Aware Quantization for Post-Training LLM Weight Compression
par: Yu, Xiaoming, et autres
Publié: (2026)

Random Conditioning with Distillation for Data-Efficient Diffusion Model Compression
par: Kim, Dohyun, et autres
Publié: (2025)

Understanding Emergent Misalignment via Feature Superposition Geometry
par: Minegishi, Gouki, et autres
Publié: (2026)

Emergence of Frontier Superposition: Möbius attractor and Cascade Supervision
par: Gu, Hongyu, et autres
Publié: (2026)

Accurate Differential Operators for Hybrid Neural Fields
par: Chetan, Aditya, et autres
Publié: (2023)

Comparative Evaluation of Applicability Domain Definition Methods for Regression Models
par: Khurshid, Shakir, et autres
Publié: (2024)

D-QRELO: Training- and Data-Free Delta Compression for Large Language Models via Quantization and Residual Low-Rank Approximation
par: Li, Junlin, et autres
Publié: (2026)

Neural Control Variates with Automatic Integration
par: Li, Zilu, et autres
Publié: (2024)

Superposition Yields Robust Neural Scaling
par: Liu, Yizhou, et autres
Publié: (2025)

Federated Two Stage Decoupling With Adaptive Personalization Layers
par: Zhu, Hangyu, et autres
Publié: (2023)

UA-PDFL: A Personalized Approach for Decentralized Federated Learning
par: Zhu, Hangyu, et autres
Publié: (2024)

Spectral Superposition: A Theory of Feature Geometry
par: Ivanov, Georgi, et autres
Publié: (2026)

Geometric Limits of Knowledge Distillation: A Minimum-Width Theorem via Superposition Theory
par: Sarkar, Nilesh, et autres
Publié: (2026)

SeedLM: Compressing LLM Weights into Seeds of Pseudo-Random Generators
par: Shafipour, Rasoul, et autres
Publié: (2024)

Large Language Model Compression with Global Rank and Sparsity Optimization
par: Zhou, Changhai, et autres
Publié: (2025)

Enhancing Delta Compression in LLMs via SVD-based Quantization Error Minimization
par: Xiong, Boya, et autres
Publié: (2025)

Breaking the Compression Ceiling: Data-Free Pipeline for Ultra-Efficient Delta Compression
par: Wang, Xiaohui, et autres
Publié: (2025)