Topics
Results for “compression”
36 results
- G-STAR: Group-Scaled Token Capacity for Deeply Compressed Autoregressive Generation — 2 sources · Seen · 46 points
- JuliusBrussee/caveman — GitHub Trending · Seen
- Neural Compression for Scientific Particle Data: Perspectives on How Learning Can Help — arXiv AI Papers · Seen
- SELD: Detecting Silent Evidence Loss in Context Compression beyond the Answer — arXiv AI Papers · Seen
- Compress any LLM 3x in one click, near-lossless, in under 50 MB RAM — Hacker News · Seen · 21 points
- ranxianglei/billion-context — GitHub Trending · Seen
- ReelCache: An Efficient KV Cache Compression and Management System for Multi-Round Video Generation — arXiv AI Papers · Seen
- ISTA-DASLab/DeepSeek-V4.1-Flash-1017-GSQ-RCO-Coder-GGUF — Hugging Face Models · Seen
- Compression of Emotion in Language Model Representations — arXiv AI Papers · Seen
- How Small Can We Go? Calibrating Activation-Cache Compression for SAE Training — arXiv AI Papers · Seen
- lz4/lz4 — GitHub Trending · Seen
- LowRAM: Low-Rank Keys and Amortised Values for Composable KV Cache Compression at Test-Time — arXiv AI Papers · Seen
- T$^2$-Splat: Adaptive Topology Mesh Splatting with Texture Residuals — arXiv AI Papers · Seen
- LZ4 1.11 Introduces Multi-Threaded Compression for Liblz4, Other Gains — Hacker News · Seen · 4 points
- LZAV 7.1: Enhanced Compression Ratio, Speed, and C/C++ Pointer Compliance — Hacker News · Seen · 3 points
- Nibnib: Image compression with extreme decompression speed — Hacker News · Seen · 186 points
- Kernel-based Interpolatory Methods for Context Compression of Attention Mechanisms — arXiv AI Papers · Seen
- H.264 Is Magic — Hacker News · Seen · 516 points
- SAC-GCL: Structure-Aware Compression for Graph Contrastive Learning — arXiv AI Papers · Seen
- GSVC: Efficient Video Representation and Compression Through 2D Gaussian Splatting — arXiv AI Papers · Seen
- ResCom: Activation Residual Compensation for Large Language Model Quantization — arXiv AI Papers · Seen
- Linear-Time Context Compression with State-Space Encoding — arXiv AI Papers · Seen
- Every Weight Is a Path of Shared Discrete Decisions — Hugging Face Papers · Seen · 23 points
- How close can pure functional programming get to C? — 2 sources · Seen · 277 points
- Bye Bye Brotli? Google's Compression Library Is a Security Mess — Hacker News · Seen · 20 points
- How Specializing the Vocabulary of SLMs Can Cut Inference Cost by 77% in Real Applications — Hugging Face Blog · Seen
- Benchmarking Model Compression Techniques Across Heterogeneous Edge Platforms: A Predictive Maintenance Use Case for Rotating Machinery — arXiv AI Papers · Seen
- MoEZip: Routing-Aware KV Cache Compression for Sparse Mixture-of-Experts LLMs — arXiv AI Papers · Seen
- headroomlabs-ai/headroom — GitHub Trending · Seen
- Embeddings Are Overpriced? A Frozen 16 MB Vocabulary for Small and Ternary Language Models — arXiv AI Papers · Seen
- ECHOES: Seeing Beyond the Current Image via Recursive History Compression and Privileged On-Policy Distillation — arXiv AI Papers · Seen
- Rewinding the exponential: How we keep hardware usage down — Cloudflare Blog · Seen
- ZXC: Lossless compression built for ultra-fast decode — Lobsters · Seen · 9 points
- PRIME: Frequency-Aware Unified Pruning and Merging for MoE Compression — arXiv AI Papers · Seen
- lyogavin/airllm — GitHub Trending · Seen
- I guess I'm using WebP now? — Hacker News · Seen · 140 points