Hyun, Jeongseok

js-hyun's activity

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

upvoted 2 papers 10 days ago

HySparse: A Hybrid Sparse Attention Architecture with Oracle Token Selection and KV Cache Sharing

Paper • 2602.03560 • Published 13 days ago • 43

OmniSIFT: Modality-Asymmetric Token Compression for Efficient Omni-modal Large Language Models

Paper • 2602.04804 • Published 12 days ago • 46

upvoted a paper 19 days ago

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

Paper • 2601.17367 • Published 23 days ago • 33

upvoted a paper 22 days ago

VideoMaMa: Mask-Guided Video Matting via Generative Prior

Paper • 2601.14255 • Published 27 days ago • 15

upvoted 2 papers 23 days ago

SAMTok: Representing Any Mask with Two Words

Paper • 2601.16093 • Published 25 days ago • 41

HERMES: KV Cache as Hierarchical Memory for Efficient Streaming Video Understanding

Paper • 2601.14724 • Published 26 days ago • 74

upvoted 4 papers about 1 month ago

Watching, Reasoning, and Searching: A Video Deep Research Benchmark on Open Web for Agentic Video Reasoning

Paper • 2601.06943 • Published Jan 11 • 212

VideoAuto-R1: Video Auto Reasoning via Thinking Once, Answering Twice

Paper • 2601.05175 • Published Jan 8 • 36

GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization

Paper • 2601.05242 • Published Jan 8 • 226

K-EXAONE Technical Report

Paper • 2601.01739 • Published Jan 5 • 91

upvoted 3 papers about 2 months ago

End-to-End Test-Time Training for Long Context

Paper • 2512.23675 • Published Dec 29, 2025 • 24

NVIDIA Nemotron 3: Efficient and Open Intelligence

Paper • 2512.20856 • Published Dec 24, 2025 • 37

Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning

Paper • 2512.20848 • Published Dec 23, 2025 • 38

upvoted 4 papers 2 months ago

QwenLong-L1.5: Post-Training Recipe for Long-Context Reasoning and Memory Management

Paper • 2512.12967 • Published Dec 15, 2025 • 108

Sliding Window Attention Adaptation

Paper • 2512.10411 • Published Dec 11, 2025 • 20

ReVSeg: Incentivizing the Reasoning Chain for Video Segmentation with Reinforcement Learning

Paper • 2512.02835 • Published Dec 2, 2025 • 10

Qwen3-VL Technical Report

Paper • 2511.21631 • Published Nov 26, 2025 • 155

upvoted 3 papers 3 months ago

Infinity-RoPE: Action-Controllable Infinite Video Generation Emerges From Autoregressive Self-Rollout

Paper • 2511.20649 • Published Nov 25, 2025 • 48

LongVT: Incentivizing "Thinking with Long Videos" via Native Tool Calling

Paper • 2511.20785 • Published Nov 25, 2025 • 185

VisMem: Latent Vision Memory Unlocks Potential of Vision-Language Models

Paper • 2511.11007 • Published Nov 14, 2025 • 15