ResearchAlpha Leak

Conferences Topics Top Authors Rankings Browse All

Home/Authors/Xinyin Ma

Xinyin Ma

Topic trends: 32,543 papers · similarity ≥ 0.4 · year ≥ 2024 · Data sourced from Semantic Scholar

34,598 papers | Abstracts: 31,650 (91.5%) | Citations: 34,598 (100.0%) | arXiv: 26,074 (75.4%)

Built: Feb 15, 2026, 12:40 AM AMS

11

papers

1,829

total citations

papers (11)

LLM-Pruner: On the Structural Pruning of Large Language Models

NEURIPS 2023arXiv

DepGraph: Towards Any Structural Pruning

DeepCache: Accelerating Diffusion Models for Free

Structural Pruning for Diffusion Models

NEURIPS 2023arXiv

dKV-Cache: The Cache for Diffusion Language Models

NEURIPS 2025arXiv

Thinkless: LLM Learns When to Think

NEURIPS 2025arXiv

TinyFusion: Diffusion Transformers Learned Shallow

Collaborative Decoding Makes Visual Auto-Regressive Modeling Efficient

VeriThinker: Learning to Verify Makes Reasoning Model Efficient

NEURIPS 2025arXiv

Diffusion Model is Effectively Its Own Teacher

Auto-Controlled Image Perception in MLLMs via Visual Perception Tokens