Soochow University

university

https://www.suda.edu.cn/

Activity Feed Request to join this org

AI & ML interests

None defined yet.

Papers

OmniSelect: Dynamic Modality-Aware Token Compression for Efficient Omni-modal Large Language Models

MemReread: Enhancing Agentic Long-Context Reasoning via Memory-Guided Rereading

View all Papers

SUDA 's papers 6

OmniSelect: Dynamic Modality-Aware Token Compression for Efficient Omni-modal Large Language Models

SUDA

Soochow University

Submitted by

Baibei Ji

MemReread: Enhancing Agentic Long-Context Reasoning via Memory-Guided Rereading

SUDA

Soochow University

Submitted by

Zecheng Tang

Elastic Attention: Test-time Adaptive Sparsity Ratios for Efficient Transformers

SUDA

Soochow University

Submitted by

Zecheng Tang

MemoryRewardBench: Benchmarking Reward Models for Long-Term Memory Management in Large Language Models

SUDA

Soochow University

Submitted by

Zecheng Tang

LongRM: Revealing and Unlocking the Context Boundary of Reward Modeling

SUDA

Soochow University

Submitted by

Zecheng Tang

Revisiting Long-context Modeling from Context Denoising Perspective

SUDA

Soochow University