zunhaisu

zunhai's activity

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

upvoted a paper 2 days ago

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus

Paper • 2608.12149 • Published 8 days ago • 30

authored 3 papers 5 days ago

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus

Paper • 2608.12149 • Published 8 days ago • 30

SnapMLA: Efficient Long-Context MLA Decoding via Hardware-Aware FP8 Quantized Pipelining

Paper • 2602.10718 • Published Apr 28

XStreamVGGT: Extremely Memory-Efficient Streaming Vision Geometry Grounded Transformer with KV Cache Compression

Paper • 2601.01204 • Published Jan 3

submitted a paper to Daily Papers 6 days ago

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus

Paper • 2608.12149 • Published 8 days ago • 30

updated a model 7 days ago

startlux-models/Massive-Activations-HLA

Text Generation • Updated 7 days ago

updated a collection 7 days ago

Massive Activations in Hybrid Linear Attention Models

Collection

Official PAS, ISP, gating-ablation, and scale-study checkpoints for Massive Activations in Hybrid Linear Attention Models. • 10 items • Updated 7 days ago

updated a model 7 days ago

startlux-models/gdn-1.3b-isp-hybrid-3to1-50b

Text Generation • 1B • Updated 7 days ago • 183

published a model 7 days ago

startlux-models/gdn-1.3b-isp-hybrid-3to1-50b

Text Generation • 1B • Updated 7 days ago • 183

updated a model 7 days ago

startlux-models/gdn-gatedfa-340m-isp-hybrid-3to1-10b

Text Generation • 0.4B • Updated 7 days ago • 182

published a model 7 days ago

startlux-models/gdn-gatedfa-340m-isp-hybrid-3to1-10b

Text Generation • 0.4B • Updated 7 days ago • 182