Papersby GlyphRDMA Point-to-Point Communication for LLM Systems research paper by Perplexity, 2025
Perplexity · Inference and efficiency · Oct 31, 2025 · 16 citations · 8 upvotes
What it shows
TransferEngine provides a uniform interface for flexible point-to-point communication in large language models, supporting disaggregated inference, reinforcement learning, and Mixture-of-Experts routing across different hardware.
By Nandor Licker, Kevin Hu, Vladimir Zaytsev and 1 more · arXiv 2510.27656 · PDF · Code
Hugging Face's summary; not yet checked by hand.
Other inference and efficiency papers
TopicEfficient 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 Berkeley9,965 citations72 upvotesRead →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 University5,830 citations16 upvotesRead →Group Sequence Policy OptimizationGroup Sequence Policy Optimization (GSPO) is a reinforcement learning algorithm that improves training efficiency and performance of large language models by using sequence-level importance ratios and operations.
Alibaba (Qwen)814 citations323 upvotesRead →Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse AttentionNSA, a trainable sparse attention mechanism, enhances long-context modeling efficiency without sacrificing performance, achieving improvements in speed and accuracy over full attention models.
DeepSeek596 citations157 upvotesRead →SmolVLM: Redefining small and efficient multimodal modelsSmolVLM, a series of compact multimodal models, achieves high performance with minimal GPU memory usage, making efficient deployment on mobile and edge devices possible.
Hugging Face330 citations212 upvotesRead →Chronos-2: From Univariate to Universal ForecastingChronos-2, a pretrained model with a group attention mechanism, achieves state-of-the-art performance in zero-shot univariate, multivariate, and covariate-informed forecasting tasks.
Amazon276 citations28 upvotesRead →
Amazon276 citations28 upvotesRead →About this paper
- Citations
- 16, 4 influential · Semantic Scholar
- Upvotes
- 8 · Hugging Face
- Lab
- Perplexity · on Companies · on Acquisitions · on awards · on Attested · on Rounds · on Mainstage · on labs · on lawsuits · on Leaders · on Allied · on Paydays · on problems · on quotes · on Releases · on repos · on salaries · on socials · on standards · on startups · on TechConf · on terms · on Titlewave · on Hotlist
Changes
Influential citations rose from 3 to 4.
Citations rose from 13 to 16.
Influential citations rose from 2 to 3.
Citations rose from 10 to 13.
A first count of 2 influential citations.
A first count of 10 citations.
Added to the list.