curius graph

all pages

showing 143351-143400 of 200407 pages (sorted by popularity)
2404.09529.pdf
Fixing DPO but I have a dinner reservation … – Kyunghyun Cho
What makes an image realistic?
1910.10683
333078981_693988129081760_4712707815225756708_n.pdf
Full Stack Optimization of Transformer Inference: a Survey
2403.14123
2106.09685
1911.02150
2305.13245
2307.08691
Stanford CRFM
2302.01318
Efficiently Scaling Transformer Inference
2306.14048
Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference
2407.15891
[2412.02252] Compressing KV Cache for Long-Context LLM Inference with Inter-Layer Attention Similarity
[2412.14838] DynamicKV: Task-Aware Adaptive KV Cache Compression for Long Context LLMs
[2502.11089] Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention
[2502.13189] MoBA: Mixture of Block Attention for Long-Context LLMs
[2410.02367] SageAttention: Accurate 8-Bit Attention for Plug-and-play Inference Acceleration
[2101.01321] I-BERT: Integer-only BERT Quantization
[2405.17849] I-LLM: Efficient Integer-Only Inference for Fully-Quantized Low-Bit Large Language Models
MaroonLLM Maybe MaroonLongLM: SelfExtend LLM Context Window Without Tuning
2210.17323
torch.Tensor.expand — PyTorch 2.7 documentation
Chebyshev Approximation and How It Can Help You Save Money, Win Friends, and Influence People - Jason Sachs
Understanding KL Divergence. A guide to the math, intuition, and… | by Aparna Dhinakaran | Towards Data Science
Star Attention: Efficient LLM Inference over Long Sequences
2310.01889
2305.19370
2502.13189
2502.12085
DeepSpeed Ulysses: System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models
RetroInfer: A Vector-Storage Approach for Scalable Long-Context LLM Inference
2402.19481
Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding
pdf
Striped Attention: Faster Ring Attention for Causal Transformers
The Race to Build a Distributed GPU Runtime | Voltron Data
Adaptive Classifier-Free Guidance via Dynamic Low-Confidence Masking
Dream 7B | HKU NLP Group
Wide-In, Narrow-Out: Revokable Decoding for Efficient and Effective DLLMs
LoongTrain: Efficient Training of Long-Sequence LLMs with Head-Context Parallelism
2405.07719
Overleaf Example
TokenRing: An Efficient Parallelism Framework for Infinite-Context LLMs via Bidirectional Communication
Context Parallelism for Scalable Million-Token Inference
Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding