Search Results - Boxing An
-
1
Batch-Max: Higher LLM Throughput using Larger Batch Sizes and KV Cache Compression
Published 2024Fuente: arXivTipo de material: PreprintAcceso al recurso -
2
On the importance of Data Scale in Pretraining Arabic Language Models
Published 2024Fuente: arXivTipo de material: PreprintAcceso al recurso -
3
Integral Transformer: Denoising Attention, Not Too Much Not Too Little
Published 2025Fuente: arXivTipo de material: PreprintAcceso al recurso -
4
BOSCH: Black-Box Binary Optimization for Short-Context Attention-Head Selection in LLMs
Published 2026Fuente: arXivTipo de material: PreprintAcceso al recurso -
5
A High‐Performance User‐Defined XFEM‐Based Cohesive Zone Modeling Method for Mixed‐Mode Fracture Problems
Published 2025Fuente: Wiley Open AccessTipo de material: Artículo Open AccessAcceso al recurso -
6
Thinking Long, but Short: Stable Sequential Test-Time Scaling for Large Reasoning Models
Published 2026Fuente: arXivTipo de material: PreprintAcceso al recurso -
7
Order-lifted data inversion/retrieval method of neighbor cells to implement general high-order schemes in unstructured-mesh-based finite-volume solution framework
Published 2024Fuente: arXivTipo de material: PreprintAcceso al recurso -
8
Single Parent Family: A Spectrum of Family Members from a Single Pre-Trained Foundation Model
Published 2024Fuente: arXivTipo de material: PreprintAcceso al recurso -
9
DPEC: Dual-Path Error Compensation Method for Enhanced Low-Light Image Clarity
Published 2024Fuente: arXivTipo de material: PreprintAcceso al recurso -
10
Do Robot Snakes Dream like Electric Sheep? Investigating the Effects of Architectural Inductive Biases on Hallucination
Published 2024Fuente: arXivTipo de material: PreprintAcceso al recurso -
11
ReGLA: Refining Gated Linear Attention
Published 2025Fuente: arXivTipo de material: PreprintAcceso al recurso -
12
Nested-ReFT: Efficient Reinforcement Learning for Large Language Model Fine-Tuning via Off-Policy Rollouts
Published 2025Fuente: arXivTipo de material: PreprintAcceso al recurso -
13
GRPO-$λ$: Credit Assignment improves LLM Reasoning
Published 2025Fuente: arXivTipo de material: PreprintAcceso al recurso -
14
EPAS: Efficient Training with Progressive Activation Sharing
Published 2026Fuente: arXivTipo de material: PreprintAcceso al recurso -
15
Exploring the compactness of $α$ cluster in $^{16}$O nuclei with relativistic $^{16}$O+$^{16}$O collisions
Published 2024Fuente: arXivTipo de material: PreprintAcceso al recurso -
16
AdpQ: A Zero-shot Calibration Free Adaptive Post Training Quantization Method for LLMs
Published 2024Fuente: arXivTipo de material: PreprintAcceso al recurso -
17
Draft on the Fly: Adaptive Self-Speculative Decoding using Cosine Similarity
Published 2024Fuente: arXivTipo de material: PreprintAcceso al recurso -
18
Rethinking Post-Training Quantization: Introducing a Statistical Pre-Calibration Approach
Published 2025Fuente: arXivTipo de material: PreprintAcceso al recurso -
19
Sorted LLaMA: Unlocking the Potential of Intermediate Layers of Large Language Models for Dynamic Inference
Published 2023Fuente: arXivTipo de material: PreprintAcceso al recurso -
20
SCOUT: Toward Sub-Quadratic Attention via Segment Compression for Optimized Utility in Transformers
Published 2025Fuente: arXivTipo de material: PreprintAcceso al recurso