ε-coresets for attention exist of size O(√d e^{ρ+o(ρ)}/ε) for unit-norm keys/values and queries of norm ≤ρ, nearly matching the Ω(√d e^ρ/ε) lower bound.
hub Mixed citations
Fast Transformer Decoding: One Write-Head is All You Need
Mixed citation behavior. Most common role is background (67%).
90
Pith papers citing it
Background
67% of classified citations
abstract
Multi-head attention layers, as used in the Transformer neural sequence model, are a powerful alternative to RNNs for moving information across and between sequences. While training these layers is generally fast and simple, due to parallelizability across the length of the sequence, incremental inference (where such paralleization is impossible) is often slow, due to the memory-bandwidth cost of repeatedly loading the large "keys" and "values" tensors. We propose a variant called multi-query attention, where the keys and values are shared across all of the different attention "heads", greatly reducing the size of these tensors and hence the memory bandwidth requirements of incremental decoding. We verify experimentally that the resulting models can indeed be much faster to decode, and incur only minor quality degradation from the baseline.
hub tools
citation-role summary
background 17
method 6
dataset 1
citation-polarity summary
claims ledger
- abstract Multi-head attention layers, as used in the Transformer neural sequence model, are a powerful alternative to RNNs for moving information across and between sequences. While training these layers is generally fast and simple, due to parallelizability across the length of the sequence, incremental inference (where such paralleization is impossible) is often slow, due to the memory-bandwidth cost of repeatedly loading the large "keys" and "values" tensors. We propose a variant called multi-query attention, where the keys and values are shared across all of the different attention "heads", greatly
co-cited works
representative citing papers
Tensor Cache augments sliding-window attention with an eviction-fed outer-product associative memory and a training correction to improve long-context performance under bounded memory.
LLMForge is a NAS framework with Infinite-Head Attention, a Forge-Former surrogate, and Forge-DSE engine that discovers hardware-specific architectures for edge language models, yielding variants with improved accuracy, energy, or latency on different substrates.
RPA kernel for TPUs achieves 86% MBU in decode and 73% MFU in prefill on Llama 3 8B via tiling for ragged memory, fused pipelines, and specialized compilation for prefill/decode workloads.
HormoneT5 augments T5 with a hormone-inspired block that predicts six continuous emotion values and uses them to modulate responses, reporting over 85% per-hormone accuracy and human preference for emotional quality.
MMEE encodes dataflow decisions in matrix form for fast exhaustive search, delivering 40-69% lower latency and energy use than prior methods while running 64-343x faster.
SCOPE accelerates autoregressive video diffusion up to 4.73x by using a tri-modal cache-predict-recompute scheduler with Taylor extrapolation and selective active-frame computation while preserving output quality.
One of the Q, K or V weights in transformer self-attention is redundant and replaceable by the identity matrix under mild assumptions, reducing parameters by 25 percent with no loss in small-model performance.
DELTA partitions layers into full, delta, and sparse groups to select salient tokens via aggregated attention scores, matching full-attention accuracy on AIME and GPQA while cutting attended tokens up to 4.25x and achieving 1.54x speedup.
FlashAttention-3 achieves 1.5-2x speedup on H100 GPUs for attention, reaching 740 TFLOPs/s (75% utilization) in FP16 and near 1.2 PFLOPs/s in FP8 while cutting numerical error by 2.6x versus baseline FP8 attention.
Transformers and SSMs are unified through structured state space duality, producing a 2-8X faster Mamba-2 model that remains competitive with Transformers.
DeepSeek-V2 delivers top-tier open-source LLM performance using only 21B active parameters by compressing the KV cache 93.3% and cutting training costs 42.5% via MLA and DeepSeekMoE.
Griffin hybrid model matches Llama-2 performance while trained on over 6 times fewer tokens and offers lower inference latency with higher throughput.
Medusa augments LLMs with multiple decoding heads and tree-based attention to predict and verify several tokens in parallel, yielding 2.2-3.6x inference speedup via two fine-tuning regimes.
A new shared video-image tokenizer enables large language models to surpass diffusion models on standard visual generation benchmarks.
Speculative sampling accelerates LLM decoding 2-2.5x by letting a draft model propose short sequences that the target model scores in parallel, then applies modified rejection sampling to keep the exact target distribution.
Speculative decoding accelerates exact sampling from large autoregressive models by 2-3x on T5-XXL by running smaller approximation models in parallel to propose token sequences that the large model then verifies in batches while preserving the original output distribution.
DREAM-S combines neural architecture search, target-aware supernet training, and attention-entropy-guided distillation to accelerate speculative decoding in VLMs, reporting up to 3.85x speedup over standard methods.
BlockBatch is a training-free framework that coordinates multiple block-size branches via token merging and synchronization to reduce denoising NFEs by 26.6% and achieve 1.33x speedup in dLLM inference.
ArborKV uses search-structure awareness to evict low-reuse KV states in Tree-of-Thoughts inference, delivering up to 4x memory savings with near-full accuracy retention.
ObjectCache enables KV cache storage in object storage via layerwise retrieval and custom scheduling, adding 5.6% latency for 64K contexts over local DRAM on a 100 Gbps RoCE cluster.
LRCP prunes visual tokens in LVLMs by scoring projection residuals onto a PCA-estimated low-rank subspace, achieving 88.9% image token reduction with 94.7% performance retention and 87.5% video reduction with 97.8% accuracy retention.
Attention to goal tokens declines in multi-turn LLM interactions while residual representations often retain decodable goal information, and the gap between these predicts whether goal-conditioned behavior survives.
ScaleSearch optimizes block floating point scales via fine-grained search to cut quantization error by 27% for NVFP4, improving PTQ by up to 15 points on MATH500 for Qwen3-8B and attention PPL by 0.77 on Llama 3.1 70B.
citing papers explorer
-
FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision
FlashAttention-3 achieves 1.5-2x speedup on H100 GPUs for attention, reaching 740 TFLOPs/s (75% utilization) in FP16 and near 1.2 PFLOPs/s in FP8 while cutting numerical error by 2.6x versus baseline FP8 attention.
-
Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality
Transformers and SSMs are unified through structured state space duality, producing a 2-8X faster Mamba-2 model that remains competitive with Transformers.
-
DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
DeepSeek-V2 delivers top-tier open-source LLM performance using only 21B active parameters by compressing the KV cache 93.3% and cutting training costs 42.5% via MLA and DeepSeekMoE.
-
Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient Language Models
Griffin hybrid model matches Llama-2 performance while trained on over 6 times fewer tokens and offers lower inference latency with higher throughput.
-
Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads
Medusa augments LLMs with multiple decoding heads and tree-based attention to predict and verify several tokens in parallel, yielding 2.2-3.6x inference speedup via two fine-tuning regimes.
-
$\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
π₀ is a vision-language-action flow model trained on diverse multi-platform robot data that supports zero-shot task performance, language instruction following, and efficient fine-tuning for dexterous tasks.
-
LightTransfer: Your Long-Context LLM is Secretly a Hybrid Model with Effortless Adaptation
LightTransfer identifies lazy layers in LLMs like LLaMA and replaces their attention with streaming attention to form hybrid models, delivering up to 2.17x throughput with under 1.5% drop on LongBench and strong results on reasoning benchmarks.
-
StarCoder 2 and The Stack v2: The Next Generation
StarCoder2-15B matches or beats CodeLlama-34B on code tasks despite being smaller, and StarCoder2-3B outperforms prior 15B models, with open weights and exact training data identifiers released.
-
KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache
KIVI applies asymmetric 2-bit quantization to KV cache with per-channel keys and per-token values, reducing memory 2.6x and boosting throughput up to 3.47x with near-identical quality on Llama, Falcon, and Mistral.
-
EAGLE: Speculative Sampling Requires Rethinking Feature Uncertainty
EAGLE resolves feature-level uncertainty in speculative sampling via one-step token advancement, delivering 2.7x-3.5x speedup on LLaMA2-Chat 70B and doubled throughput across multiple model families and tasks.
-
DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models
DeepSeekMoE 2B matches GShard 2.9B performance and approaches a dense 2B model; the 16B version matches LLaMA2-7B at 40% compute by using fine-grained expert segmentation plus shared experts.
-
Gemma: Open Models Based on Gemini Research and Technology
Gemma introduces open 2B and 7B LLMs derived from Gemini technology that beat comparable open models on 11 of 18 text tasks and come with safety assessments.
-
Yi: Open Foundation Models by 01.AI
Yi models are 6B and 34B open foundation models pretrained on 3.1T curated tokens that achieve strong benchmark results through data quality and targeted extensions like long context and vision alignment.
-
Gemma 2: Improving Open Language Models at a Practical Size
Gemma 2 models achieve leading performance at their sizes by combining established Transformer modifications with knowledge distillation for the 2B and 9B variants.
-
ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools
GLM-4 models rival or exceed GPT-4 on MMLU, GSM8K, MATH, BBH, GPQA, HumanEval, IFEval, long-context tasks, and Chinese alignment while adding autonomous tool use for web, code, and image generation.
-
A Survey on Efficient Inference for Large Language Models
The paper surveys techniques to speed up and reduce the resource needs of LLM inference, organized by data-level, model-level, and system-level changes, with comparative experiments on representative methods.