Shuohuan Wang

wangshuohuan's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a paper 3 days ago

Multi-View MoE: Letting Every Router Shape Expert Computation

Paper • 2610.40577 • Published 9 days ago • 2

authored 7 papers 14 days ago

RESUME: Recurrent State Updates from Motion and Residual Signals for Efficient Video Language Modeling

Paper • 2609.39563 • Published Sep 30

When Does Better Acceptance Pay? A Controlled Study of Speculative Decoding

Paper • 2610.24993 • Published 18 days ago • 2

ECHO: Efficient Causal Self-Recursion Beyond Auxiliary Multi-Token Prediction

Paper • 2610.24766 • Published 21 days ago • 15

Autonomy-of-Heads: Data-Free Sparse Attention from Frozen Query-Key Geometry

Paper • 2608.06849 • Published Aug 7

CoVisco: Codec-Native Vision Encoder with Native Token Compression for Unified Image-Video Understanding

Paper • 2609.39924 • Published Sep 30 • 3

Aggregating Visual Information with Optimal Transport for VideoLM Token Compression

Paper • 2608.20473 • Published Aug 20

ConSA: Controllable Sparsity in Hybrid Attention via Learnable Allocation

Paper • 2606.18056 • Published Jun 16

upvoted a paper 16 days ago

ECHO: Efficient Causal Self-Recursion Beyond Auxiliary Multi-Token Prediction

Paper • 2610.24766 • Published 21 days ago • 15

authored a paper 5 months ago

Memento: Reconstruct to Remember for Consistent Long Video Generation

Paper • 2606.14667 • Published Jun 12 • 19

upvoted a paper 5 months ago

Memento: Reconstruct to Remember for Consistent Long Video Generation

Paper • 2606.14667 • Published Jun 12 • 19

authored 9 papers 5 months ago

Sparse Growing Transformer: Training-Time Sparse Depth Allocation via Progressive Attention Looping

Paper • 2603.23998 • Published Apr 16 • 1

Learning to Generate via Understanding: Understanding-Driven Intrinsic Rewarding for Unified Multimodal Models

Paper • 2603.06043 • Published Mar 6

Blink: Dynamic Visual Token Resolution for Enhanced Multimodal Understanding

Paper • 2512.10548 • Published May 23

V-ITI: Mitigating Hallucinations in Multimodal Large Language Models via Visual Inference-Time Intervention

Paper • 2512.03542 • Published Dec 3, 2025

Elastic MoE: Unlocking the Inference-Time Scalability of Mixture-of-Experts

Paper • 2509.21892 • Published May 11

BeamLoRA: Beam-Constraint Low-Rank Adaptation

Paper • 2502.13604 • Published Jun 12, 2025

Clip-Tuning: Towards Derivative-free Prompt Learning with a Mixture of Rewards

Paper • 2210.12050 • Published Oct 21, 2022

Galileo at SemEval-2020 Task 12: Multi-lingual Learning for Offensive Language Identification using Pre-trained Language Models

Paper • 2010.03542 • Published Oct 7, 2020

ERNIE 3.0: Large-scale Knowledge Enhanced Pre-training for Language Understanding and Generation

Paper • 2107.02137 • Published Jul 5, 2021