Efficient Attention

Follow this topic Ask about this topic

Momentum

15 papers in the last four weeks, up 114% on the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 96

Oct 7, 2026cs.DS

Attention via Black-Box Vector Search

Sparse attention mechanisms estimate attention over

n

tokens using a small subset of keys. Many existing approaches use maximum inner product search (MIPS) to retrieve the heaviest keys, which motivates the following question: given black-box access to a MIPS oracle, how many keys must be retrieved to output an

\varepsilon

-accurate attention estimate? We answer this question by unifying prior approaches through the framework of priority sampling. With a single MIPS index, we show that

Θ(\sqrt{n}/\varepsilon)

retrieved keys are both sufficient and necessary. With

Θ(\log n)

indices, we give an algorithm that retrieves only

O(\log n+1/\varepsilon^2)

keys and prove that this is near-optimal. More generally, we design algorithms that establish a smooth tradeoff between the number of MIPS indices and number of retrieved keys. We then show that if we allow augmentation of keys and queries, we can bypass the above lower bounds: there exists a simple priority-sampling estimator using a single MIPS index and

O(1/\varepsilon^2)

retrieved keys. When integrated into LLM inference, our algorithms outperform top-

k

and sampling approaches used in prior work and yield attention approximation that scales favorably to long contexts.

Stepan Zharkov, Krish Singal, Ashwin Padaki +1

University of Pennsylvania

Efficient Attention Sparse Attention LLM Inference Efficiency

Sep 30, 2026cs.LG

LampAttention: Look-Ahead Mixed-Precision FlashAttention for Dedicated Accelerators

While most attention logits can be computed in low precision without degrading numerical stability, current attention kernels fail to exploit this phenomenon. We introduce a novel hardware-algorithm co-design in the form of mixed-precision FlashAttention. Our method accumulates key-query products and evaluates their exponentials in 8-bit formats, then adaptively identifies sensitive sub-blocks and recomputes them in 16-bit formats. We propose the specifications for a dedicated accelerator capable of executing this pipeline efficiently. Simulated experiments with Qwen3 and Gemma 3 show that rerouting a selective minority of sub-blocks to high precision is sufficient to recover the baseline model performance.

Stanislav Budzinskiy, Marian Gloser, Tolunay Yilmaz +5

Faculty of Mathematics University of Vienna, Austria · Huawei Heisenberg Research Center, Munich, Germany · Huawei Technologies Co. Ltd

AI Accelerator Inference LLM Inference Acceleration Efficient Attention Attention Mechanisms

Sep 30, 2026cs.LG

Switching Linear Attention

Designing expressive sequence layers with efficient inference remains a central challenge in modern machine learning. Standard softmax attention achieves excellent sequence modeling performance through rich nonlinear token interactions, but it requires a key-value cache that grows linearly with sequence length, limiting its scalability. Linear attention enables efficient recurrent computation with a constant memory footprint, yet its reduced expressivity often yields inferior modeling performance. We introduce Switching Linear Attention (SwiLA), a novel sequence layer that bridges this gap by enhancing representational capacity while retaining the fixed-size recurrent state of linear attention. We derive the SwiLA recurrence from the test-time regression framework, casting the state update rule as online expectation-maximization in a mixture of linear regressions model. At test time, each output dimension dynamically selects among multiple linear attention components based on the input. Across associative recall, in-context language learning, and language modeling benchmarks, SwiLA shows strong performance and narrows the gap to softmax attention, even surpassing it in several settings.

Hyun Dong Lee, Xavier Gonzalez, Nicolas Zucchet +3

Stanford University · Unconventional AI. Work done at Stanford University.

Efficient Attention Linear Attention

Sep 28, 2026cs.LG

SANTA++: Sampling Attention through Representative Keys

Attention often concentrates on a small subset of tokens in the context, but which subset matters changes from one query to the next. To exploit this changing structure, we introduce SANTA++, a training-free stochastic attention method that uses representative keys for memory-efficient selection without scanning the entire key-value (KV) cache. Cached keys are organized into teams, and the query scores one representative from each team to decide which teams to sample. We compute exact attention scores within the sampled teams and reweight each team's contribution by the inverse of its inclusion probability. This importance sampling correction estimates attention over the full cache, with a sampling budget that lets us trade memory reads for accuracy. Remarkably, with 32 or 64 sampled teams, SANTA++ uses 16% to 22% of dense attention's KV reads and retains 94% to 99% of the dense-attention baseline's scores on LongBench v2 and HELMET's retrieval-augmented generation subset, and 85% to 91% on RULER, with Qwen2.5-7B-Instruct at 32K context. With 31 sampled teams, our GPU implementation delivers a

1.69\times

attention speedup over the dense FlashAttention baseline at 32K context. By reducing the number of cache entries read, SANTA++ in principle complements architectures with compressed KV representations, such as multi-head latent attention. Our kernels are available at: https://github.com/OPUSLab/santapp-kernel-demo.git.

Kyle Lee, Christian Z. Pratt, Ruoyu Fang +4

Department of Electrical and Computer Engineering University of California, Santa Barbara, CA, USA · Flucta, San Francisco, CA, USA

Memory-Efficient Inference LLM Inference Acceleration Efficient Attention

Sep 28, 2026cs.LG

When Can Attention Heads Be Statically Defined?

Some attention heads learn similar patterns across inputs. Reusing these patterns could reduce training cost by avoiding repeated query-key score computation and softmax. Through controlled pretraining comparisons, we identify Selective Attention Freezing (SAF), which selects heads with low attention-pattern variance and replaces their attention weights with fitted post-softmax means halfway through training. We represent these fixed patterns with absolute-position and relative-distance preferences, reducing storage from quadratic to linear in sequence length. A fused kernel reconstructs the patterns and executes ordinary-attention and replaced heads together. At matched training-token budgets, replacing 25% of attention heads gives 1.056x faster post-replacement optimiser updates at 124M parameters and 4K context, with a 0.77% perplexity increase. At 1B and 8K context, post-replacement updates are 1.068x faster on four GPUs including communication, with a 0.51% perplexity increase. The resulting models also accelerate long-input finetuning and causal prefill. After associative-recall adaptation, the 124M model with 25% replacement generalises to more key-value pairs at a fixed length better than ordinary attention and two pruning controls.

Weixian Waylon Li, Yintao Tai, Marcio Fonseca +1

University of Edinburgh, United Kingdom · Chamber of Deputies, Brazil

Attention Head Analysis Efficient Attention Attention Head Specialization

Sep 28, 2026cs.LG

Broken Symmetry in BF16 Attention: Why FlashAttention Gradients Blow Up Late in Training

BF16 is now standard in large-scale pretraining, including in fused attention kernels such as FlashAttention, and these kernels are widely trusted. When we used FlashAttention-3 to pretrain a 450M-parameter transformer on 50B tokens, however, we ran into a problem: training was healthy for 25B tokens, then the gradient norm grew a thousandfold and the loss ended 0.2 nats above FP32 attention, without a single NaN. Recomputing the attention backward of just two layers in FP32 removes almost all of the excess gradient. Part of the cause is known: a fused multiply-add in the forward softmax, so far treated as an extreme-input NaN case and never fixed in FlashAttention-3. Repairing it stops the blow-up, but the query gradient is still wrong by more than its own size, and training still drives attention logits to thousands of times their size under accurate gradients. The remaining error comes from a broken conservation law. The softmax score gradient sums to zero along every row, which makes the query gradient blind to where the keys sit as a group; rounding it to BF16 leaves a small nonzero sum that leaks the mean key into the gradient, and the leak grows exactly as late training makes keys large and attention sharp. We introduce GProj (gauge projection), which restores the zero sum after the cast with two rank-one corrections per row. It cuts the remaining median query/key gradient errors from 219%/13% to 0.34%/0.37%, on par with FP32 attention, for 4.7% more time per training step. In matched from-scratch runs it trains to the same loss as FP32 attention, while FlashAttention-3 and key smoothing both destabilize.

Junlin Chen, Daize Dong, Huanwei Di +9

Rutgers University · Carnegie Mellon University · Oracle +2

Softmax Attention Efficient Attention Attention Mechanisms

Sep 27, 2026cs.LG

FoldAttention: Declared-Reference Softmax for Fast Decode and Deterministic Backward

Autoregressive decode repeatedly streams a growing KV cache, making attention a major cost at long context. Existing high-performance kernels use online softmax, which discovers a row's normalization reference as it scans keys. Earlier contributions therefore remain provisional and may require rescaling. We argue that the reference need not be discovered: softmax is invariant to a common shift, so the reference only has to keep the weights in range. We present FoldAttention, an additive formulation of softmax attention that fixes a finite reference

Z_i

before scanning the KV cache. Each weight

2^{s_{ij}-Z_i}

is then final when computed, so contributions add across disjoint key ranges and their quotient equals softmax attention in real arithmetic. We use this property to develop two techniques for Hopper decode: (1) final weights gate key and value reads before the bytes are fetched, and a per-call depth

T

cuts keys below

2^{-T}

while keeping their mass, and (2) additive partials compose split KV and shared-prefix cascades without rescaling. On H100 at

T=16

, FoldAttention decodes seven real-model generations 1.36-2.30

\times

faster than the fastest BF16 baseline, and up to 3.09

\times

faster across MHA and GQA shapes, at an error within 1.5% of the lowest BF16 error on six of the seven; reading every key, it is 1.14-1.30

\times

faster at matched error. We validate on Qwen3-8B that a whole decode step is up to 1.46

\times

faster while likelihood and long-context accuracy match those under BF16 kernels. The same principle makes the backward deterministic: CTAs round bounded partial gradients onto an integer grid declared before the reduction and add them in any order. FoldAttention thereby removes the determinism tax: its deterministic backward is up to 1.84

\times

faster than deterministic FlashAttention-3/4 and 1.05

\times

faster than the fastest nondeterministic kernel.

Sriman Achanta

Virginia Commonwealth University

Softmax Attention LLM Inference Acceleration Efficient Attention Autoregressive Decoding

Sep 25, 2026cs.LG

Benchmarking Attention for Tabular Foundation Models

Tabular in-context learners such as TabPFN, Mitra, or ConTextTab rely on alternating row and column attention over 2D sequences of latent embeddings. These attention patterns differ markedly from the one-dimensional case in language models: row attention involves longer sequences while column attention operates on much shorter ones, and the strided memory layout of tabular data makes producing contiguous tensors costly. Moreover, the hidden dimensions used in current models are small compared to recent language models. Yet efficient attention has been studied mostly for one-dimensional sequences, leaving the two-dimensional tabular setting unexplored. To this end, we create a reproducible benchmarking setup and study the unique characteristics of tabular attention across several backends -- Torch SDPA (efficient and cuDNN), FlashAttention-2/3/4, and the inference-only backends vLLM and SageAttention -- measuring forward and backward throughput across realistic tabular shapes on three GPU generations (A100, H100, B200). We find that the optimal backend choice differs between column and row attention and varies across hardware as well as model specifics: While the FlashAttention implementations tailored for each GPU generation perform overall best, they are at times outperformed by CuDNN in the case of column attention at longer sequences with cross-over points depending on the head dimension. Among inference-only backends, SageAttention performs well for row attention and large sequences beyond 16,k rows. Our reproducible benchmark lays the foundation for future improvements to table-native attention. The self-contained benchmarking and evaluation code is openly available at: https://github.com/SAP-samples/tabular-attention-benchmark

Maximilian Schambach, Clemens Biehl, Sam Thelin

SAP SE, Germany

Tabular Foundation Models Efficient Attention Tabular ICL

Sep 23, 2026cs.CL

Attention Routing Stabilizes Early: Working-Set Inference for Recurrent Language Models

Recurrent-depth language models, such as looped Transformers, repeatedly apply shared network blocks to refine latent representations without generating explicit intermediate reasoning tokens. However, each step recomputes full attention over the entire context, repeating costly global routing. We study how attention routing evolves across recurrent depth and find a consistent separation in convergence timescales: attention support and distributions stabilize substantially earlier than hidden states and attention outputs. This suggests two stages of recurrent inference: early discovery of a sparse working set, followed by representation refinement over largely stable routing support. Motivated by this finding, we introduce WISE (Working-set Inference with Support Exploitation), a training-free method that uses unrestricted attention during early recurrent steps to discover a block-structured working set, then reuses its support in later steps while keeping attention weights and recurrent refinement dynamic. Controlled interventions show that multi-step discovery yields more effective working sets than first-step selection, and that support reuse better preserves model behavior than more restrictive forms of attention reuse. Across multi-hop QA benchmarks, WISE largely preserves full-attention performance. Matched context-scaling experiments reveal an increasingly favorable quality-efficiency tradeoff as routing support becomes sparser with longer contexts. A sparse-attention implementation achieves up to a 1.76x late-step attention speedup over native FlashAttention at 4K context. Code: https://github.com/tbn5pj/WISE_code.

Ke Wan, Chen Chen

Department of Computer Science University of Virginia Charlottesville, VA, USA

Long-Context Language Model Inference Recurrent Transformers Efficient Attention Efficient Language Model Inference Block-Sparse Attention

Sep 21, 2026cs.LG

FlashBoB: I/O-Efficient Exact Backward-over-Backward for Softmax Attention

Transformer models built on the attention mechanism have become a central building block in modern deep learning, yet softmax attention remains a major bottleneck for long-context workloads. While FlashAttention makes the forward and first backward passes I/O-efficient, it does not support backward-over-backward (BoB), which enables exact differentiation through the backward pass for applications such as second-order optimization, test-time training, gradient-based memory, and meta-learning. Existing BoB implementations either materialize large intermediate tensors or exhaust GPU memory at long sequence lengths. We present FlashBoB, an exact, I/O-efficient algorithm for BoB in softmax attention that keeps computation within on-chip tiles and avoids all

N \times N

intermediate tensors, where

N

is the sequence length. The key insight is a hierarchical affine structure in the softmax double backward: two row-wise scalars determine all outputs through affine transformations. This yields a two-pass schedule with bounded on-chip static random-access memory (SRAM) usage and minimal off-chip high-bandwidth memory (HBM) traffic. FlashBoB achieves

Θ(N^2 d^2/M)

HBM traffic (

d

is the head dimension and

M

is the memory size) and, within the standard FlashAttention-style score-recomputation model, matches the inherited large-cache lower bound for exact forward attention. Empirically, it scales exact attention BoB to

N=262\text{K}

on a single A100 80GB GPU, where prior PyTorch exact baselines fail by

N=16\text{K}

, and is up to

6.3\times

faster than FlashBack. These results make exact second-order attention practical at long-context sequence lengths where prior implementations cannot run efficiently.

Anthony Givans, Michael Crawshaw, Mingrui Liu

Softmax Attention Automatic Differentiation Efficient Attention

Sep 16, 2026cs.LG

Reaching Every Position Without Searching: Rotating Sparse Wiring on the Hypercube as a Substitute for Attention

Attention pays, at every layer and for every input, the cost of searching for whom to connect. We ask how far one can get with wiring that is fixed, sparse, and simply rotated from layer to layer. Treating the

n

positions of a sequence as the vertices of a

\log_2 n

-dimensional hypercube and connecting each position, at layer

\ell

, to its neighbour along dimension

\ell \bmod \log_2 n

, information from every position reaches every other in

\log_2 n

layers with

2n

links per layer instead of

n^2

. On a synthetic task that is unsolvable unless all positions are reached, this rotation matches all-to-all wiring at

1/32

of the links, while the same sparse pattern held fixed across layers fails; what matters is that every dimension is touched, not the order. On character-level language modelling of a public corpus (the first

12

M characters of enwik8), a hybrid that keeps two attention layers among sixteen sparse ones reaches

0.06

bits-per-character lower held-out loss than a fully attentive model of the same width at the same step budget (three seeds each, no overlap), with

1/7

of the links,

42\%

fewer parameters, and

2.4\times

less wall-clock time; the purely rotated schedule is level with the hybrid. The same ordering holds on a second corpus of mixed Japanese, English and code, where the gap widens to

0.16

. The usable learning-rate window is four to eight times wider than attention's on both. We also report what did not work - learned coordinates, and a "dynamics" variant whose apparent gains turned out to be an artefact of a saturated kernel - and the measurement discipline (frozen corpus, full-coverage evaluation, seed spread as the bar for ranking) that we found necessary to say anything at all at this scale.

Yoshiaki Takashita

School of Law, Waseda University

Efficient Attention Language Modeling Sparse Attention

Sep 14, 2026cs.AI

Rethinking Heterogeneous System Disaggregation for Subquadratic Attention

Frontier language models are more aggressively using subquadratic attention to reduce the memory footprint and compute requirements during inference while still delivering frontier accuracy. While existing systems make dense attention-centric disaggregated serving decisions, we show that disaggregating inference around the unique arithmetic intensity and memory footprint of subquadratic attention LLMs can achieve significant throughput and energy efficiency gains on emerging DRAM-based and SRAM-only heterogeneous systems. We introduce SQD (SubQuadratic Disaggregation), a fine-grained heterogeneous disaggregation scheme that splits decode by quadratic and subquadratic attention rather than by operator type, and that applies across subquadratic attention variants. For sparse attention LLMs, we disaggregate decode into top-k selection, which must index through the full KV, and top-k attention plus FFN, which have static memory footprints. For linear and sliding-window attention LLMs, we disaggregate decode into dense attention layers and subquadratic attention layers plus FFN. In an adjusted 8xB200 heterogeneous system proxy, we observe average tokens/J improvements of 53% on GLM 5.2, 31% on Nemotron 3 Ultra, and 56% on Gemma 4 31B over the strongest GPU-only baselines. In an analytical model of a Rubin plus LPX system with fixed power budgets, we observe 1.2x to 1.5x tighter achievable latencies and up to 3.6x higher throughput over the best baseline of attention-FFN disaggregation. Our experiments also reveal architectural insights on chip and interconnect provisioning for next-generation heterogeneous systems serving subquadratic attention.

Arya Tschand, Yaosheng Fu, Vikram Sharma Mailthody +6

Harvard University Cambridge, Massachusetts, USA · NVIDIA Santa Clara, California, USA

Disaggregated LLM Serving LLM Inference Acceleration Heterogeneous Computing Efficient Attention

Sep 8, 2026cs.AR

Grouped Value Attention: Efficient KV Caching via On-Demand Key Reconstruction

The KV cache is a primary bottleneck for Transformer decoding: its memory footprint and cache-read traffic grow with sequence length. Grouped-query attention (GQA) reduces this cost by sharing key-value heads, but still stores both a key and a value at every step. We introduce Grouped Value Attention (GVA), which stores grouped values and reconstructs content keys with a learned linear map. At inference, the map can be absorbed into the query, eliminating the need to materialize content keys in the intended decode path. A small shared decoupled RoPE channel retains positional information through a separately cached positional key. For the configurations studied, this representation reduces persistent cache scalars by approximately 45-47% relative to matched GQA. At the 350M-parameter scale with 30B FineWeb-Edu tokens, the 16-dimensional positional variant reaches 44.18 average accuracy across five tasks, compared with 44.36 for GQA and 43.88 for MLA. These results demonstrate near-GQA benchmark accuracy with a more compact cache representation. To translate this compact representation into faster autoregressive inference, we have developed custom decoding kernels and are currently evaluating their end-to-end inference performance with an open-source release planned soon.

Vishesh Tripathi, Abhay Kumar, Ramsha Khan

FrontiersMind

Efficient Transformer Inference Grouped-Query Attention KV Caching Efficient Attention KV-Cache Compression

Sep 8, 2026cs.LG

Nyström Attention Matches Full Attention for Cross-Sectional Stock Prediction

MASTER's inter-stock multi-head attention -- the module responsible for modeling cross-sectional stock relationships -- accounts for 42.5% of model parameters and 25% of predictive value. We systematically decompose this module and uncover a surprising structure: the learned attention is near-uniform (perplexity 278/300), yet forcing exact uniformity eliminates all cross-sectional discrimination. Spectral analysis resolves this paradox: the deviation from uniformity is low-rank (effective rank ~65, top-10 modes capture 96.5% of energy), explaining why sparse approximations consistently fail while Nystrom low-rank attention (m=32 landmarks) matches full O(N^2) attention at O(mN) cost -- certified equivalent via TOST at both N=300 (5 seeds, Rank IC p=0.003) and N=800 (10 seeds, Rank IC p=0.034). Additional findings include: (i) attention anti-correlates with return similarity (Spearman rho = -0.614; on the industry-labeled subset, -0.645 unconditionally and -0.627 after controlling for industry, beta, and volatility), suggesting complementarity-seeking rather than correlation mining; (ii) all graph-based alternatives degrade performance, with hard masking worse than complete module removal; and (iii) at N ~ 3,500 with adapted architectures, no cross-stock module (GCN, Nystrom, or MASTER-style pipeline) significantly outperforms a per-stock LSTM baseline (n=4 seeds), indicating that the benefits observed at smaller scales do not trivially transfer. These results establish that the inter-stock attention's value resides in a compressible, dynamic, near-global redistribution that rewards low-rank approximation but resists sparsification.

Kunhan Guo

The University of Hong Kong

Low-Rank Attention Financial Forecasting Time Series Forecasting Efficient Attention Attention Mechanisms

Sep 7, 2026cs.CL

RouteRelay: Event-Triggered Cross-Layer Route Reuse for Efficient Dynamic Sparse Attention

Dynamic sparse attention reduces long-context prefill cost by routing each query chunk to a small set of key chunks at every Transformer layer. The sparse attention kernel avoids most token interactions, but the router still rebuilds a chunk--chunk score matrix layer after layer, even when the selected routes change little. We introduce RouteRelay, a router-agnostic method that reuses only route metadata across depth while continuing to compute attention with the current layer's queries, keys, and values. Anchor layers perform full routing. Intermediate layers rescore the previous top-

k

route and a compact sentinel set of near-miss and randomly probed chunks. A query row is rerouted only when a sentinel challenges its weakest selected chunk. We give a top-

k

stability condition, a probabilistic bound on missed challengers, and a row-selective GPU execution design. In a reproducible empirical evaluation, RouteRelay retains at least 99.99% route recall while rerouting 25.0%, 55.4%, and 78.2% of rows under low, moderate, and high cross-layer drift, respectively. Across routing scales, RouteRelay retains 100.0% recall while evaluating 38.4--51.6% of full-routing score pairs as the key-chunk count grows from 128 to 1024. Its unfused CPU execution remains slower than dense matrix multiplication, exposing row compaction and ledger updates as the main kernel-engineering targets.

Bin Li, Sisi Liu, Chenyang Hu +3

Department of Computer Science, Xiamen University of Technology · School of Mathematics and Computer Science, Fuzhou University

Adaptive Model Routing Efficient Attention Sparse Attention

Sep 7, 2026cs.CL

CEDAR: Error-Bounded Residual Routing for Efficient Long-Context Attention

Post-hoc sparse attention accelerates long-context prefill by routing each query to a small set of token-level interactions. Hard selection, however, assigns zero probability to every omitted chunk: a routing miss cannot be recovered, and a fixed expansion budget spends the same work on easy and ambiguous queries. We introduce Coarse-to-fine Error-aware Dynamic Attention Routing (CEDAR), a coarse-to-fine method that keeps the language model frozen while preserving global coverage. Each semantic chunk contributes a cheap key--value summary to a residual attention path; chunks with high estimated approximation error are then expanded to exact token attention. Exact and summarized contributions are combined in a single softmax normalization, so refinement replaces, rather than duplicates, coarse evidence. We derive an output-error bound governed by within-chunk key/value dispersion and use it to allocate a variable refinement budget. A controlled clustered-attention study shows that residual summaries reduce reconstruction error by more than 98% relative to hard dropping at equal exact-chunk budgets. Experiments on long-context benchmarks demonstrate that CEDAR recovers most of the quality lost by hard sparse routing while maintaining approximately

3\times

kernel speedup at 128K context.

Siyu Li, Dong Wang, Jie Zhou +3

School of Computer Science & Technology, Shanghai University · School of Computer Science & Technology, East China Normal University · School of Computer Engineering, Jiangsu Ocean University

Long-Context Language Modeling Efficient Attention Sparse Attention

Sep 3, 2026cs.LG

Hardware-Aware FP4 FlashAttention-4

Blackwell's 4-bit floating-point (FP4) tensor cores do not automatically make attention faster because softmax conversion and on-chip dependencies dominate once its matrix products shrink. We address this with \emph{Direct-P} for noncausal inference and a causal path that passes the forward quantization directly into backward. Direct-P maps scores directly to FP4 probabilities and reaches up to 2.13

\times

the bfloat16 (BF16) forward throughput on an NVIDIA GB200. The causal path reconstructs probabilities from saved quantized queries and keys and uses 8-bit floating-point (FP8) gradient operands, accelerating a complete single-GPU 8-billion-parameter update by up to 1.14

\times

. Matched distributed training retains FP8 probabilities and values; every tested MXFP4 probability/value training trajectory diverges.

Robert Hu

GRAPHCORE RESEARCH

Mixed-Precision Quantization GPU Kernel Optimization Transformer Attention Efficient Attention Attention Quantization Attention Mechanisms

Aug 31, 2026hep-ex

ConCA: Concentration-Aware Channel Attention for Fine-Grained Visual Recognition

Lightweight channel attention mechanisms are widely used in image classification, yet their effectiveness in fine-grained visual recognition (FGVR) remains limited. Most modules summarize each channel by global average pooling (GAP), which captures activation magnitude but ignores spatial concentration, so channels with different spatial distributions but identical means receive the same descriptor. We propose Concentration-Aware Channel Attention (ConCA), which pairs the mean with a shift-invariant negative-input entropy (NegEnt), computed via a softmax over the negated activations, forming a dual descriptor that jointly encodes magnitude and concentration. A depthwise 1-D convolutional multi-layer perceptron (MLP), whose parameter count is linear in the number of channels, maps the pair to a per-channel weight. On six fine-grained benchmarks, ConCA improves over attention-free, SE-Net, and ECA-Net baselines as well as four richer descriptor-based modules under a controlled from-scratch protocol, and it generalizes across eight backbones on iNat2021-mini. These results indicate that the channel descriptor, together with the per-channel gating that maps it to attention weights, is an important but underexplored aspect of lightweight channel attention in FGVR.

Yu-Sheng Liu, Yu-Chen Tung

Department of Physics, National Kaohsiung Normal University, Kaohsiung, 824, Taiwan

Channel Attention Fine-Grained Image Classification Efficient Attention Fine-Grained Visual Perception

Aug 27, 2026cs.LG

ClusterAttention: A training-free speedup of bidirectional attention

We introduce ClusterAttention, a general training-free speedup of bidirectional attention at large token counts. We point out two common assumptions in contemporary training-free methods; attention sparsity, and context that can be leveraged, such as structure in the input or multiple similar forward passes, and show when they fail. Our proposed method utilizes a fast attention-aware recursive clustering method, and compensation of excluded clusters through their mean. The clustering method gives power-of-two cluster sizes, allowing block-sparse attention to match dense attention in GPU throughput. On TabPFN-3 arXiv:2605.13986, a model where none of the assumptions hold, ClusterAttention is to our knowledge the first method to provide a substantial speedup over the default attention, while consistently keeping over 99% of its accuracy. On the largest dataset from the TALENT benchmark suite, it makes processing of the training dataset close to 8x faster at nearly 11x attention speedup. ClusterAttention is also competitive with domain-specific methods, while avoiding any of the domain-specific engineering. On video-generation with Wan 2.1-T2V-14B arXiv:2503.20314 it produces output closer to dense attention at a larger speedup (1.8x vs 1.4x) than SVOO arXiv:2603.18636, a leading method in this domain, with both evaluated without offline calibration.

Kasper Nordenram, Amelie Dittmann

Independent researcher

Efficient Attention Attention Mechanisms

Aug 13, 2026cs.CV

SCOPE: Subspace Clustering with Online Per-Head Top-K Estimation for Sparse Video Attention

Diffusion Transformers (DiTs) incur quadratic self-attention cost over spatiotemporal tokens. Existing training-free sparse attention methods often construct sparse masks from block-level or cluster-level proxy scores, which can obscure fine-grained differences among keys and miss high contribution keys under aggressive sparsity. Moreover, such proxy scores may yield overly concentrated softmax distributions, causing Top-

p

to retain too few keys for some query clusters. Although a fixed Top-

k

minimum alleviates this failure mode, a shared value cannot adapt to variations across heads and inputs. To address both limitations, we propose SCOPE, a training-free sparse attention framework that combines 3D-RoPE-aligned key subspace clustering with online per-head Top-

k

estimation for efficient video-DiT inference. SCOPE partitions post-RoPE keys into temporal, height, and width subspaces, clusters them independently, and aggregates the corresponding centroid scores through lookup tables to obtain per key proxy scores for each query cluster. Building on existing hybrid Top-

p

/fixed Top-

k

selection, SCOPE derives a head-specific Top-

k

value online by averaging the initial retained key counts within each head, weighted by query cluster size, and selects additional keys only for query clusters whose initial retained key counts fall below this value. Sparse attention is then computed over the selected original keys and values. Across six model--task configurations, SCOPE consistently outperforms existing training-free baselines in both fidelity and latency, achieving up to a

1.99\times

end-to-end speedup on 720p HunyuanVideo with

28.46

dB PSNR relative to dense attention.

Qi Zhao, Qirui Li, Hanlin Tang +10

Zhejiang University · Alibaba Group

Video Diffusion Models Diffusion Transformer Clustering Efficient Attention Sparse Attention Diffusion Model Inference Acceleration

Aug 12, 2026cs.NE

Lapis: Laplacian Spiking Attention via First-Spike Timing and Membrane Leakage

Self-attention has become central to spiking vision transformers, yet its query-key scoring is still largely inherited from dense networks. Existing spiking variants either simplify dot product scoring or replace it with discrete operators, but spike timing, the native variable of a spiking network, does not directly define how tokens are related. We propose Lapis, a spiking attention mechanism that scores each token pair by the L1 distance between its query and key first-spike latency vectors under time-to-first-spike coding, and maps this distance to an affinity through a Laplacian kernel. The kernel's exponential decay matches the impulse response of a leaky integrate-and-fire membrane, so the accumulated latency difference determines the decay of a membrane trace, while row normalization reduces to a bit shift under power-of-two rounding. Scoring therefore needs only subtraction, absolute value, and accumulation, and removes all multiplication between query and key channels. Under a matched backbone and training schedule, Lapis reaches 96.56% top-1 accuracy on CIFAR-10, within 0.53 points of dot-product scoring. On ImageNet-1K, it reduces the estimated arithmetic energy of the attention path by 14.5x relative to dense dot-product attention. The deployed 6-bit model attains 83.25% top-1 accuracy at an estimated arithmetic energy of 3.28mJ per image.

Kaiwen Tang, Jiaqi Zheng, Zixuan Zhu +3

School of Computing, National University of Singapore · 2Sea AI Lab · 4Shanghai Advanced Research Institute, Chinese Academy of Sciences +2

Spiking Transformers Efficient Attention Spiking Neural Networks Attention Mechanisms

Aug 12, 2026cs.CL

Hybrid Gated Attention

Gated attention is an effective approach to mitigate attention sinks and enhance the representational capacity of attention. To further extend its effectiveness-efficiency Pareto frontier, we propose a Hybrid Gated Attention (HyGA) framework that contains three types of gating strategies. Specifically, these gates leverage diverse information from multiple stages of attention, and collaboratively build element-wise/head-wise gating from multiple perspectives, capturing intra-head and cross-head information interactions. Through our hybrid gating components, HyGA could provide multi-source modulation signals, enabling more comprehensive control over information flow and improving the representational capacity of attention. We also introduce low-rank matrix decomposition and learnable attention sink to further enhance training efficiency and stability. In experiments, we evaluate HyGA on widely-used benchmarks based on different backbones. The experimental results show that our HyGA comprehensively improves both training loss and various downstream performances compared with Gated attention. HyGA has also been verified to achieve the best performance at different computation costs, with comprehensive model analyses for better understanding. The proposed HyGA sheds light on a more effective, efficient, and stable attention mechanism.

Zekun Zhou, Ruobing Xie, Lanrui Wang +1

Tencent Hunyuan · Peking University

Low-Rank Matrix Decomposition Gated Attention Transformer Attention Efficient Attention Attention Sinks Hybrid Attention

Aug 11, 2026cs.LG

Efficient Reinforcement Learning for Long-Horizon Tool-Use Agentic Tasks

Long-horizon tool-using agents must reason over user goals, domain policies, tool calls, simulator state, and delayed verifiable rewards. Reinforcement learning (RL) is a natural fit for this setting, but multi-turn on-policy rollouts create long contexts, while model-specific attention layers may require custom masks and learned sink normalization. We present SINKFLEX-RL, a modular training system for RL in dual-control tool-use environments. The system combines a Gymnasium-compatible environment wrapper, a VERL-style rollout dataflow, group-relative policy optimization without a separate value model, and a sink-aware FlexAttention path designed to preserve model-specific sink scaling under causal and sliding-window masks. In a preliminary Tau2Bench retail run, validation reward (mean@1) rises from 0.25 early in training to

0.44

later in the observed training window, while training-score and trajectory-reward proxies also trend upward. In a fixed-configuration memory benchmark, the optimized attention path reduces peak VRAM from 28.06GB to 22.52GB at 4096 tokens, a

19.7\%

reduction, and runs the measured 8192-token configuration using

25.53

~GB where the eager baseline runs out of memory. These results illustrate the value of integrating environment interfaces, RL dataflow, and attention-kernel design for memory-feasible long-horizon agent training.

Zelei Cheng, Amritansh Mishra, Sambit Sahu +1

AI Foundations, Capital One

Efficient Attention Long-Horizon Agent Tasks RL for LLM Agents RL for Tool Use

Aug 5, 2026cs.LG

Training-Free Hashing-Based Attention via Binary Principal Components

Long-context large language models (LLMs) are increasingly deployed in real-world applications, yet self-attention remains a major efficiency bottleneck -- especially during decoding -- due to the necessity of repeatedly processing ever-growing key-value (KV) caches. Existing sparse attention reduce computation by attending to fewer KV pairs, but often suffer from substantial accuracy degradation, require additional training, or rely on expensive hashing. In this work, we present BinaryPC, a training-free, data-aware hashing-based sparse attention for long-context LLMs. BinaryPC constructs compact binary hash codes and corresponding hash function by computing binary principal components of data. Unlike Locality-Sensitive Hashing (LSH) with data-independent random projections or learned non-linear hashing methods, BinaryPC constructs binary codes that explicitly preserve the structural information of data without requiring gradient-based training. Comprehensive experiments across multiple model families and long-context benchmarks show that BinaryPC preserves accuracy relative to full attention while achieving superior performance among sparse and hashing-based baselines. On modern GPUs, BinaryPC improves end-to-end decoding throughput by 3.56

\times

over the FlashAttention kernel. Our code is available at https://github.com/yudaohai666/BPC.

Daohai Yu, Zhanpeng Zeng, Keyu Chen +6

Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University, 361005, P.R. China. · Tencent YouTu Lab, Shenzhen, China · Sino-Russian Research Center for Digital Economy.

Long-Context Language Modeling LLM Inference Acceleration Efficient Attention Sparse Attention

Aug 3, 2026cs.AI

LongCat Sparse Attention: Taming the Lightning via Streaming-aware Hierarchical Cross-Layer Indexing

DeepSeek Sparse Attention (DSA) enables efficient long-context modeling through its Lightning Indexer. However, practical deployment remains constrained by the indexer's expensive

O(L^2)

scoring overhead and the hardware-inefficient, discontinuous memory-access patterns induced by its outputs. To address these system-level bottlenecks, we introduce LongCat Sparse Attention (LSA), a hardware-algorithm co-designed framework comprising three complementary and orthogonal strategies: (1) Streaming-Aware Indexing, which selectively converts scattered KV entries into hardware-aligned contiguous layouts to enable coalesced HBM access; (2) Cross-Layer Indexing, which amortizes indexing overhead by reusing the results produced by a single layer across consecutive layers, supported by cross-layer distillation; and (3) Hierarchical Indexing, which adopts a coarse-to-fine scoring scheme to progressively narrow the candidate set for each query, thereby substantially reducing indexing computation. Extensive scaling experiments, ranging from 69B-A3B to 560B-A27B models, demonstrate that LSA consistently achieves performance on par with full attention across both general-purpose and long-context benchmarks. Moreover, LSA supports native training with context lengths of up to one million tokens and underpins the development of LongCat-2.0 (1.6T-A48B). To facilitate further research, we also introduce and open-source LongCat-Flash-Lite-Sparse (69B-A3B), which integrates LSA into LongCat-Flash-Lite and incorporates an updated long-context training corpus.

Wen Zan, Jiaqi Zhang, Jianchao Tan +11

Meituan LongCat Team

Long-Context Modeling LLM Inference Acceleration Efficient Attention Sparse Attention

Jul 30, 2026cs.LG

S-CEReBrO: Breaking the Memory Barrier in Continuous EEG Monitoring

Foundation models offer a promising paradigm for Electroencephalography (EEG) analysis, leveraging generalizable representations from vast unlabeled datasets. Yet, Transformer-based architectures face a critical bottleneck: global attention mechanisms couple the attention memory state to the signal duration, causing memory overflow during continuous monitoring. To address this, we introduce S-CEReBrO (Streaming CEReBrO), an evolution of the CEReBrO architecture designed for continuous monitoring. Our novel Windowed Alternating Attention mechanism factorizes attention computation into fixed-size spatiotemporal windows, so that under streaming, only the active window remains resident and the attention state is bounded independently of signal duration. Empirical scaling analysis shows that windowed alternating attention can process signals 100X longer than full self-attention and 3X longer than low-rank linear attention. Compared to low-rank linear attention on long contexts, windowed alternating attention requires 55% of the memory while increasing inference throughput by 2.1X. Pre-trained on >25,000 hours of recordings from >12,000 subjects, S-CEReBrO achieves state-of-the-art performance on 7 of 11 downstream tasks, with up to 60% fewer parameters. This work represents a significant step toward the realization of efficient, generalizable, and continuous EEG monitoring. An accompanying code repository is available. An accompanying code repository is available.

Glenn Anta Bucagu, Thorir Mar Ingolfsson, Yawei Li +1

ETH Zurich, Zurich, Switzerland · Nanyang Technological University, Singapore · University of Bologna, Bologna, Italy

Efficient Transformer Inference Electroencephalography Efficient Attention EEG Foundation Models

Jul 30, 2026cs.CL

Recall Before You Rank: Similarity-Guided Top- $K$ Reuse for Efficient Long-Context Attention

Top-

K

sparse attention reduces the cost of Softmax and value aggregation by attending to only a small subset of key--value (KV) entries. However, identifying this subset still requires scoring the current query against the full KV cache and performing global Top-

K

selection, leaving selector cost linear in context length and limiting the practical efficiency of sparse attention for long-context decoding. In this paper, we introduce ReTopK, a training-free method that accelerates dynamic Top-

K

attention by reusing historical retrieval decisions. ReTopK builds on the observation that similar queries often attend to overlapping supports and that partially overlapping supports can still preserve most of the Exact Top-

K

attention mass. For each attention head, it maintains a bounded cache of historical query--support pairs, retrieves the most similar cached queries for each new query, unions their stored supports with a recent window, and reranks only the resulting compact candidate set using exact current-query scores. A similarity-based fallback invokes full-history Exact Top-

K

when reuse is unreliable, while periodic exact refreshes limit cache drift. ReTopK retains the complete KV cache and reuses only selected indices, rather than historical scores, attention weights, or outputs. Across 16K--128K contexts, ReTopK achieves the lowest PG19 perplexity and the highest NIAH and LongBench scores among the evaluated approximate methods. At 128K with

K=512

, ReTopK incurs only a 0.50% perplexity increase over Exact Top-

K

while accelerating attention computation by

3.07\times

.

Wenshuai Yao, Wenyong Zhou, Hanyong Shao +5

School of Integrated Circuits, Peking University, Beijing, China · Department of Electrical and Computer Engineering, The University of Hong Kong, Hong Kong SAR, China · School of Integrated Circuit Science and Engineering, Beihang University, Beijing, China

Long-Context Language Model Inference Efficient Attention Sparse Attention

Jul 27, 2026cs.CL

PIVOT: Efficient Query-Group Indexing for Token-Level Sparse Attention

Token-level sparse attention, as implemented by DeepSeek Sparse Attention (DSA) in production systems, makes the downstream attention efficient but shifts the bottleneck to the indexer that feeds it. To select the top-k tokens for each query, the indexer must still score every preceding token, incurring a cost of O(L^2) per layer for a sequence of length L. We observe that this per-query scan is largely redundant: nearby queries select highly overlapping top-k tokens, and the indexer scores are long-tailed along the key axis. We exploit these properties in PIVOT, Proxy Indexing Via One full-prefix Traversal, a training-free, drop-in replacement for the DSA indexer that shares one prefix scan across a group of nearby queries. PIVOT aggregates a group into a single proxy query, performs one shared full-prefix scan to obtain a candidate set, and then selects a top-k for each query from that set. Two variants trade speed for fidelity: PIVOT-Reuse shares the proxy top-k across the group for maximum speed, whereas PIVOT-Refine re-scores the candidate set with the indexer of each query and then selects an individual top-k, matching the dense indexer at a small additional cost. A single algorithm covers both inference phases, differing only in how groups are formed: fixed-size groups of consecutive queries in prefill, and the queries decoded together in one multi-token prediction (MTP) step in decode. On DeepSeek-V3.2 and GLM-5.1 across LongBench and RULER, PIVOT matches the accuracy of the dense DSA indexer while accelerating it by up to 4x and reducing end-to-end latency by up to 1.6x at long context.

Hong Liu, Yuan Cheng, Lin Niu +5

Tencent

LLM Inference Acceleration Efficient Attention Sparse Attention

Jul 26, 2026cs.LG

A Coulomb Particle Model for Learning Kernel Attention in Transformers

Randomized features provide a scalable approximation to kernel machines, but their performance depends strongly on the choice of feature distribution. We propose a particle-based method that learns this distribution by optimizing kernel-target alignment while regularizing particles with a Riesz/Coulomb repulsive potential. The resulting Hamiltonian yields diverse, task-adaptive random features and admits a mean-field description through a McKean--Vlasov equation. We instantiate the method in linearized Transformer attention by learning positive random-feature maps in a first alignment phase, then freezing the kernel and training the remaining network parameters with cross-entropy. Experiments on synthetic classification and sentence-level benchmarks show that learned kernelized attention can improve accuracy, calibration, and robustness for several feature maps while preserving linear-attention inference complexity.

Masoud Badiei Khuzani, Sharath Honnaiah, Atiq Islam +2

eBay Inc., San Jose, CA, USA

Transformer Attention Efficient Attention Random Feature Methods Kernel Methods

Jul 22, 2026cs.LG

ELSAA: Efficient Low-Rank and Sparse Attention Approximation for Training Transformers

The quadratic

N\times N

attention score matrix remains a central obstacle to extending Transformers to longer input lengths. Existing efficient attention methods usually reduce this bottleneck by either imposing sparsity, so that each query attends to only a small subset of keys, or by using low-rank/kernel sketches, so that global interactions are compressed into a lower-dimensional representation. We propose \emph{ELSAA}, an efficient low-rank and sparse approximation of attention. Importantly, ELSAA does \emph{not} decompose the learned projection or output matrices of the Transformer into sparse and low-rank factors. Instead, after dense projections produce

Q,K,V

, ELSAA approximates the induced attention score operator itself: a sparse branch captures selected high-similarity interactions, while a low-rank branch summarizes diffuse global interactions. Since the two branches can be normalized over supports with very different denominator mass, ELSAA introduces a denominator-aware fusion term that scales the sparse branch according to its estimated attention mass relative to the low-rank branch. This gives a practical framework for constructing low-rank and sparse attention outputs without materializing the full quadratic score matrix, aiming to enable longer-context training while preserving both sharp token-level interactions and broad contextual mixing.

Mahdi Heidari, Mohammad Mahdi Rahimi, Jaekyun Moon

School of Electrical Engineering, KAIST, Daejeon 34141, Republic of Korea · Department of Electrical Engineering and Computer Science, Daegu Gyeongbuk Institute of Science and Technology (DGIST), Daegu 42988, Republic of Korea

Low-Rank Attention Transformer Attention Efficient Attention Sparse Attention Attention Mechanisms