Skip to content
Papers.

SwiftKV: Fast Prefill-Optimized Inference with Knowledge-Preserving Model Transformation research paper by Snowflake, 2024

Snowflake · Oct 4, 2024 · Inference and efficiency · 18 citations · 2 upvotes

Read on arXiv

What it shows

SwiftKV lets prompt tokens skip later layers and merges KV caches, cutting prefill cost for long-prompt workloads.

Summarised by hand from the abstract.

More from Snowflake

All 3
PaperCitations
Strategic Navigation or Stochastic Search? How Agents and Humans Reason Over Document CollectionsOn 2,250 questions over 800 PDFs, the best agents match human accuracy but lean on brute-force search rather than planning.Agents and evaluation · Mar 20261
Arctic-Embed: Scalable, Efficient, and Accurate Text Embedding ModelsOpen text embedding models from 22M to 334M parameters that led MTEB retrieval for their size at release.Retrieval and data · May 202480

Other inference and efficiency papers

Topic
PaperCitations
Efficient Memory Management for Large Language Model Serving with PagedAttentionvLLM manages the KV cache like pages of virtual memory, serving models with 2 to 4 times the throughput.UC Berkeley · Sep 20238,469
FlashAttention: Fast and Memory-Efficient Exact Attention with IO-AwarenessFlashAttention computes exact attention with far fewer GPU memory reads and writes, making long sequences faster.Stanford University · May 20225,331
DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache CompressionA 552B mixture-of-experts model with a 1M-token context, built to shrink the KV cache for agent workloads.DeepSeek · Sep 20267
About this paper
Authors
Aurick Qiao, Zhewei Yao, Samyam Rajbhandari and 1 more
arXiv
2410.03960 · PDF
Venue
Conference on Empirical Methods in Natural Language Processing
Citations
18, 1 influential · Semantic Scholar
Upvotes
2 · Hugging Face
Lab
Snowflake · on Companies · on Acquisitions · on Quarterly · on Paydays · on Releases · on TechConf

Changes

What changed
New paperAdded to the listSep 24, 2026

Sources: each lab's own papers and arXiv, with citation and upvote counts from Semantic Scholar and Hugging Face. One-line summaries are for orientation, not a substitute for the paper. Logos via logo.dev; trademarks belong to their owners.

New papers by email

Monday afternoons, only in weeks with new papers from the labs.

Double opt-in. Unsubscribe any time.