paperAug 11, 2026▲ 28
Zunhai Su, Bohan Sun, Xialie Zhuang et al.
We present the first systematic study of Massive activations (MAs) in layer-interleaved HLA LLMs and uncover two architecture-aligned morphologies: MAs consistently spike immediately before full attention layers, forming pre-attention spikes (PAS), and can persist through intervening linear attention layers, giving rise to inter-spike plateaus (ISP). As full attention becomes denser, successive PAS become increasingly connected through ISP, ultimately recovering the stable MA morphology of full attention LLMs. We establish the recurrence of this organization across five linear attention architectures, six hybridization configurations, five data domains, and representative open-source hybrid models spanning 1.2B to 397B total parameters. Controlled pretraining of GDN-based hybrids at scales up to 1.3B shows that both morphologies emerge early and respond asymmetrically to output gating: full attention output gating strongly attenuates their absolute magnitudes without eliminating their layerwise organization, whereas removing GDN gates yields comparatively modest amplification. Mechanistically, our systematic-outlier analysis supports a shared lifecycle account governed by the timing of MA cancellation. PAS follows a localized write-sink-cancel process, while the extended persistence of ISP is consistent with delayed cancellation. At the full attention limit, this account recovers the stable MA morphology characteristic of full attention LLMs. Our code is available at https://github.com/StartluxLabs/Massive-Activations-HLA.
cond-mat.dis-nn · cond-mat.stat-mechSep 23, 2026new
Qucheng Gao, Zuyi Yang, Xiao Chen
We study the nonequilibrium dynamics of a minimal recurrent transformer with $N$ normalized tokens, $Q=K=I$, and a negative value map $V=-I$. Similarity-based attention selects nearby representations, while the negative value map drives tokens away from the selected field. This feedback can continually reorganize both the representation geometry and the attention network. For $d=2$, the tokens lie on a circle, where the regular polygon is an exact fixed point. As the attention feedback strength $γ$ is increased, the polygon loses stability through a flip bifurcation, giving rise to period-two motion, chaos, and cluster-exchange or cluster-flip states. Despite this temporal complexity, attention remains diffuse as $N\to\infty$ at finite fixed softmax sharpness $β$. Attention condensation instead emerges in the scaling regime $β\sim N^2$. In the hard-routing limit, repulsive updates amplify local perturbations and routing-partner switches transmit them ballistically, producing an emergent butterfly cone in representation space. High-dimensional geometry provides a distinct route to localization. For $d=N\to\infty$, simulations from Gaussian initial conditions provide evidence for a condensation transition at $β=O(1)$, driven by dynamically generated finite overlap gaps. Depending on $γ$, the resulting phases include diffuse simplex-like states, consensus flips, condensed active routing with signatures of chaos, and fragmented cluster flips. These results establish temporal activity, attention condensation, and geometric clustering as distinct collective phenomena, and show that sparse attention can sustain persistent dynamics rather than freeze it.
paperJul 22, 2026▲ 34
Junsong Chen, Jincheng Yu, Yitong Li et al.
We introduce SANA-Video 2.0, a hybrid video diffusion transformer instantiated at 5B and 14B scales under a unified architecture. Designed to generate high-quality video up to 720p on a single GPU, SANA-Video 2.0 matches full-softmax video DiTs in quality while retaining the favorable long-sequence scaling of linear attention. To avoid quadratic attention throughout, Hybrid Linear-Softmax Attention combines gated linear attention for O(N)-dominated mixing with periodic gated-softmax anchors at a 3:1 ratio, restoring the full-rank token interactions that pure linear attention lacks. To propagate these refreshed representations across depth, Block Attention Residuals (AttnRes) route completed block summaries into later linear layers, enabling anchor-feature reuse and boosting deep-layer effective rank by ~12%. Through from-scratch training, SANA-Video 2.0 learns the complete hybrid directly rather than linearizing pretrained models, with reduced-resolution proxy studies establishing 25% softmax as the optimal quality-efficiency trade-off. With 40-step sampling, SANA-Video 2.0 achieves a VBench score of 84.30 in 13.2s at 480p on a single H100, remaining competitive with far larger softmax video DiTs at a fraction of the latency. Its compiled DiT forward pass is 3.2x faster than a matched full-softmax baseline at 720p/60s, a gap that expands with video duration. Furthermore, full-stack Sol-Engine optimization (kernel fusion, caching, and sparse attention) accelerates this hardware-friendly backbone by a further 3.58x, bringing the 5B pipeline to 13.06s at 720p/5s and making it 120x faster than Wan 2.2-A14B on one H100. Overall, our hybrid design recovers softmax-level expressiveness at substantially reduced cost, unlocking scalable long, high resolution video generation.
paperSep 13, 2026▲ 36
Xingyang Li, Dongyun Zou, Shining Zhang et al.
Diffusion Transformers deliver state-of-the-art video generation, but their long spatiotemporal sequences make attention the dominant deployment cost, and a deployable low-bit kernel must be accurate and fast. Accuracy is limited by outliers: a block's quantization scale is set by its largest entries, leaving typical entries confined to a narrow range of representable values. Prior work smooths queries and keys, but value outliers follow no fixed channel or spatiotemporal structure and remain the dominant source of output error. Speed is limited by softmax: low-bit Tensor Cores accelerate only the two matrix multiplications, so the high-precision exponential between them becomes the longest pipeline stage on datacenter GPUs. We propose VC-Attention, a training-free low-bit attention framework that addresses both by pairing Value smoothing with a fused probability Cast. V-Smooth reorders value tokens by lightweight online clustering, so the tokens in a hardware block quantize well together. It quantizes only the residual after subtracting the block mean, and restores that mean from the row sum the online softmax already maintains. ExpCast-FP8 maps log-domain scores directly to E4M3 probability codes with one fused multiply-add, eliminating the FP32 exponential and the format conversion. We implement VC-Attention for B200, B300, H200, RTX PRO 6000, and RTX 5090. Across Wan2.2, LongCat-Video, HunyuanVideo-1.5, and MiniMax-H3, VC-Attention improves fidelity over low-bit baselines, speeds up the attention kernel over BF16 FlashAttention-4 by 1.46-1.59x on datacenter Blackwell and Hopper and by 2.3-3.6x on workstation cards, and generates a clip 1.13-1.19x and 1.36-1.70x faster end to end.
quant-ph · cs.LGAug 11, 2026new
Eric A. F. Reinhardt, Adam J. Hauser
The attention mechanism forms the foundation of many modern AI models such as the Transformer. In one subclass of problems where attention is used, inputs and outputs are bound to the probability simplex so that all outputs sum to one. In this setting, softmax attention admits an exact, component-by-component quantum realization. Attention scores are Hadamard-test statistics on block-encoded projections of amplitude-encoded inputs. The exponential softmax is the interior of a cosine-squared family generated by Born-rule measurement under an exact bijection, whose boundary expresses sparse attention with exact zeros at finite parameter values. The softmax temperature is a repetition count where post-selected measurement rounds realize discretized inverse temperature exactly. Value aggregation is a deterministic column-loading channel that dilates the column-stochastic value matrix. The gated residual is the preparation angle of a single ancilla, with the additive identity at a mixing angle of π/2. Every learnable parameter is a rotation-gate angle. The composed layer is exact in the infinite-shot limit with one measure-and-reload step per attention score; a fully-coherent variant is ε-approximate via quantum singular value transformation in the infinite depth limit. The algebraic core is machine-checked in Lean 4.
paperNeurIPS 20252025342 citations
Zi-Han Qiu, Zekun Wang, Bo Zheng et al.
Gating mechanisms have been widely utilized, from early models like LSTMs and Highway Networks to recent state space models, linear attention, and also softmax attention. Yet, existing literature rarely examines the specific effects of gating. In this work, we conduct comprehensive experiments to systematically investigate gating-augmented softmax attention variants. Specifically, we perform a comprehensive comparison over 30 variants of 15B Mixture-of-Experts (MoE) models and 1.7B dense models trained on a 3.5 trillion token dataset. Our central finding is that a simple modification-applying a head-specific sigmoid gate after the Scaled Dot-Product Attention (SDPA)-consistently improves performance. This modification also enhances training stability, tolerates larger learning rates, and improves scaling properties. By comparing various gating positions and computational variants, we attribute this effectiveness to two key factors: (1) introducing non-linearity upon the low-rank mapping in the softmax attention, and (2) applying query-dependent sparse gating scores to modulate the SDPA output. Notably, we find this sparse gating mechanism mitigates 'attention sink' and enhances long-context extrapolation performance, and we also release related $\href{https://github.com/qiuzh20/gated_attention}{codes}$ and $\href{https://huggingface.co/QwQZh/gated_attention}{models}$ to facilitate future research.
paperSep 25, 2026▲ 5
Jingze Shi, Zhangyang Peng, Xianduo Li et al.
FullAttn often assigns negligible normalized mass to much of the causal score space, yet dense kernels execute the complete post-score path after forming each QK tile. We introduce MALA, a fused attention primitive that preserves score access to every legal causal interaction and uses normalized contribution to allocate post-score computation. Forward uses its evolving online-softmax normalizer, while backward reuses the finalized normalizer to derive nested retained support using only standard attention state. A common tolerance governs training and inference, allowing for adaptive retention of the work. MALA reduces low-contribution post-score computation. A matched-work study at 8K isolates the benefit of distribution-adaptive allocation: under exactly matched total post-score work, MALA approaches a per-instance reference-mass oracle, with mean omitted mass of 0.0188% versus 0.0182%. Across context lengths from 1K to 32K tokens, the same tolerance maintains low output and gradient errors relative to the reference. Across a broader controlled associative-recall comparison, MALA closely tracks FullAttn as context grows, reaching 89.67% accuracy at 8K compared with 89.97% for FullAttn. In an attention-operator benchmark at 128K tokens with tensor parallelism, MALA reduces forward and backward latency during training by 2.2x and 3.0x and decoding latency during inference by 1.6x relative to FullAttn. Across scaling-law training from 0.6B to 14B parameters, MALA closely tracks FullAttn in perplexity while reducing total training FLOPs. The resulting 14B models and 32B models from separate continued training achieve comparable knowledge, reasoning, and long-context retrieval scores to FullAttn. These results indicate that allocating post-score computation according to normalized attention contributions can retain the evaluated capabilities of FullAttn while reducing attention computation.
cs.CV · cs.LGSep 29, 2026new
Joseph Metcalfe, Sara Sharifzadeh, Fabio Caraffini
The landscape of satellite imagery time series datasets and boundary-pushing architectures for cropland segmentation has never been richer. However, in this gold rush, important truths are being missed on both fronts, as a drive for the most novel concepts or the largest datasets pushes finer details to the side. In this paper, we present our hybrid transformer-convolutional model, Cropland Parallel Attention and Refinement Network for Segmentation (PAtteRNS), the first model to use self-attention mechanisms separately for each of the temporal, spectral, and spatial aspects of Sentinel-2 multispectral SITS data. To achieve fully-factorised attention in our proposed model, we introduce a novel parallel transformer architecture which significantly reduces the computational complexity of triple-factorised self-attention. We validate our architecture with an in-depth ablation study, and analyse the performance of our model against state-of-the-art crop segmentation models on multiple tile-size variants of the popular PASTIS and MTLCC datasets. Our findings show our model to outperform all others in the task of crop class segmentation, verified across multiple important segmentation metrics, with especially strong performance against compared models seen in the often under-reported parcel delineation quality, for which we use the Boundary IoU metric. We also find that flawed class groupings within datasets can have a significant negative impact on model performance, and report that alternate tile-size variants of crop segmentation datasets produce results incomparable to one-another, invalidating fair comparison between model performance when trained on different tile-sizes. Based on these findings, we suggest further work is required to standardise best practices when constructing SITS crop segmentation datasets, and to enable future dynamic-tile-sizing for ideal model performance.
paperSep 9, 2026▲ 5
Changbo Yan, Zhongbo Zhang, Zaibin Zhang et al.
3D point-cloud observations are inherently ambiguous in complex, cluttered manipulation scenes, where target objects may be partially occluded or tightly intermingled with visually similar distractors. As a result, standard 3D diffusion policies often struggle to localize and exploit task-relevant geometry as scene complexity grows. We propose Attention-DP3, a spatially object-aware 3D diffusion policy that injects object-level geometric cues via attention while keeping the DP3 diffusion backbone unchanged. Our pipeline performs open-vocabulary 2D segmentation on RGB images, then lifts predicted target masks into 3D using calibrated camera geometry to obtain object-centric geometric priors. We incorporate these cues through Tri-field Attentional Conditioning, which constructs three complementary fields: (i) a targetness field to anchor the target object, (ii) an intra-target saliency field to emphasize task-relevant geometry within the target, and (iii) a backgroundness field to suppress distractors and clutter. Experiments on Adroit, DexArt, MetaWorld, and the real-world SO101 platform show consistent improvements over DP3, achieving state-of-the-art performance across benchmarks. Notably, as distractor objects increase, DP3 drops sharply, whereas Attention-DP3 remains stable and outperforms DP3 by up to 31\% under heavy clutter. The code is publicly available at https://github.com/zhangzhongbo2213/Attention-DP3.
cs.LGAug 19, 2026new
Sotirios P. Chatzis, Loukas Papadoulas
Deep models for irregularly-sampled time series answer queries at arbitrary continuous timestamps, yet report nothing about how far each answer should be trusted. We show the attention layer itself can close that gap: with the right stochastic formulation, the pass that makes each prediction also reports, in closed form and at no extra cost, how far it should be trusted. We introduce Lévy Attention, a cross-attention operator whose output is a stochastic integral against an inhomogeneous Poisson random measure: query-key compatibilities assemble an intensity over a continuous (time x channel) index space, the measure scatters atoms under it, and the output averages an interpolated value field at those atoms. In expectation it reduces to a mollified cosine-kernel attention, so it replaces a softmax layer and trains with exact gradients.
What softmax discards, the Poisson construction preserves in closed form: the evidence $Λ_q$ (total compatibility mass) and the disagreement $\mathrm{tr}\,Σ_V(q)$ (value spread). An exact variance identity makes their combination $\hatσ(q)=\sqrt{\mathrm{tr}\,Σ_V(q)\,\varphi(Λ_q)}$ the root-mean-square deviation of the sampled operator, emitted by the deterministic pass with no trained head.
Empirically, disagreement carries the signal, while the evidence factor swings from uninformative on dense data to strongly informative on sparse. On t-PatchGNN the operator swap costs at most 5.6% accuracy against a matched control and nothing on the sparsest dataset. The free disagreement signal improves on 20-pass MC dropout across matched five-seed suites, and $\hatσ$ scales a calibrated Gaussian whose zero-sample CRPS beats a fifty-draw sampler; a split-conformal wrapper reaches nominal coverage at every level, and one pass ranks 3,383 unseen patients by trust in 1.4 seconds.
paperAug 9, 2026▲ 7
Burc Gokden
The Large Language Model from Power Law Decoder Representations (PLDR-LLM) and its attention, Power Law Graph Attention (PLGA), replace the fixed bilinear form of scaled dot-product attention (SDPA) with a learned, input-generated bilinear operator G_{LM}, built from a positive tensor A_{LM} by elementwise power laws. The architecture is fully specified, verified against pinned reference releases; claims are labeled theorem, conditional theorem, measurement, or conjecture. Unconditionally: PLGA contains SDPA exactly at G_{LM}=I; A_{LM} and A_P are strictly entrywise positive, with Perron-Frobenius structure on A_{LM}; the DAG regularizer has the NOTEARS walk-counting form and positivity obstructs exact acyclicity; and, under nonresonance (satisfied by standard rotary frequencies), a commutant criterion identifies which operators preserve relative-position dependence. An inference-collapse theorem: exact input invariance of deductive outputs collapses inference to generalized SDPA with a constant operator. Measured invariance: relative fluctuations of 10^{-6} and below; perturbation bounds quantify but do not certify cached inference; the assembled proxy misses the decoding margin. A conditional three-stage mechanism (rotary twirl, concentration, row-map contraction) is measured on a released checkpoint. Blockwise training and scoring under the global Gram are stated with explicit target exposure; on tested samples, block and sequential scoring select identical answers and agree on the published TruthfulQA probability-mass metric within 5times 10^{-5} per item. Self-organized criticality enters as a phenomenological framework with an intrinsic order parameter; open claims become falsifiable conjectures. Selected proof cores are machine-checked in Lean 4.
paperJul 2, 2026▲ 44
Xiang Hu, Xinyu Wei, Hao Gu et al.
Scaling modern large language models (LLMs) to long contexts is limited by the quadratic computation cost, and poor length extrapolation of dense attention. Chunk-wise sparse attention offers a promising alternative, but all existing methods fall short of full attention because of their inaccurate chunk selection. We propose Hierarchical Landmark Sparse (HiLS) Attention, a chunk-wise sparse attention mechanism that learns chunk selection end-to-end under the language-modeling (LM) loss. HiLS factorizes attention hierarchically: each query performs attention independently with each retrieved chunk to extract chunk-specific information, and the resulting outputs are fused according to chunk retrieval scores. By incorporating retrieval scores into the forward attention computation, HiLS optimizes them directly with the LM loss, enabling end-to-end retrieval learning and native sparse training. Experimental results show that HiLS-Attention achieves performance comparable to, and in some cases better than, full attention at in-domain context lengths. Meanwhile, HiLS-Attention extrapolates more than 64times the training context length with 90% retrieval accuracy, far beyond full attention. Moreover, existing full-attention models can be converted to HiLS-Attention with lightweight continued pretraining, preserving in-domain performance while acquiring ultra-long-context extrapolation. Together with its sparse KV access and computation, HiLS-Attention breaks the usual efficiency-performance trade-off, enabling long-context LLMs that are both more efficient and more effective on general long-context tasks than their full-attention counterparts.
paperICLR 20242024232 citations
Xiangming Gu, Tianyu Pang, Chao Du et al.
Language Models (LMs) assign significant attention to the first token, even if it is not semantically important, which is known as attention sink. This phenomenon has been widely adopted in applications such as streaming/long context generation, KV cache optimization, inference acceleration, model quantization, and others. Despite its widespread use, a deep understanding of attention sink in LMs is still lacking. In this work, we first demonstrate that attention sinks exist universally in LMs with various inputs, even in small models. Furthermore, attention sink is observed to emerge during the LM pre-training, motivating us to investigate how optimization, data distribution, loss function, and model architecture in LM pre-training influence its emergence. We highlight that attention sink emerges after effective optimization on sufficient training data. The sink position is highly correlated with the loss function and data distribution. Most importantly, we find that attention sink acts more like key biases, storing extra attention scores, which could be non-informative and not contribute to the value computation. We also observe that this phenomenon (at least partially) stems from tokens' inner dependence on attention scores as a result of softmax normalization. After relaxing such dependence by replacing softmax attention with other attention operations, such as sigmoid attention without normalization, attention sinks do not emerge in LMs up to 1B parameters. The code is available at https://github.com/sail-sg/Attention-Sink.
paperSep 10, 2026▲ 53
Zhiwei Li, Lei Zhu, Hao Gu et al.
Post-training attention sparsification reduces the quadratic cumulative attention cost of pretrained Transformers by selecting a small set of context units (tokens or blocks) for each query. Existing trainable methods usually use a lightweight selector to score context units, followed by hard Top-K selection that blocks gradients from the language modeling loss. Consequently, these methods commonly distill layer-wise dense attention distributions. Although this encourages the selector to rank context units by dense attention weights in the original model, the ranking is not directly aligned with their impact on predictions under a fixed attention budget (i.e., the number of attended context units per query), potentially wasting the limited budget on less useful units. To address this misalignment, we propose Simple Attention Sparsification (SAS), a gated sparse attention mechanism that optimizes context ranking end-to-end with the language modeling loss. The key idea is to inject the selector's continuous scores into attention logits during training, allowing the loss to update the selector through standard backpropagation. We identify several choices crucial for this simple design to work well in practice: placing the gate inside the attention softmax in log form, using normalized softmax gates to calibrate historical context against the always-retained current block, and preserving continuous selector scores so the model learns relative priorities rather than only hard selections. To support long-sequence training, we implement a memory-efficient Triton kernel that integrates SAS into FlashAttention-style computation. Across reasoning, long-context understanding, and agentic tasks, SAS consistently outperforms trainable sparse attention baselines across attention budgets, with especially large gains under tight budgets, demonstrating more effective context ranking for downstream tasks.
paperICLR 20252025124 citations
Xunhao Lai, Jianqiao Lu, Yao Luo et al.
Large language models (LLMs) encounter computational challenges during long-sequence inference, especially in the attention pre-filling phase, where the complexity grows quadratically with the prompt length. Previous efforts to mitigate these challenges have relied on fixed sparse attention patterns or identifying sparse attention patterns based on limited cases. However, these methods lacked the flexibility to efficiently adapt to varying input demands. In this paper, we introduce FlexPrefill, a Flexible sparse Pre-filling mechanism that dynamically adjusts sparse attention patterns and computational budget in real-time to meet the specific requirements of each input and attention head. The flexibility of our method is demonstrated through two key innovations: 1) Query-Aware Sparse Pattern Determination: By measuring Jensen-Shannon divergence, this component adaptively switches between query-specific diverse attention patterns and predefined attention patterns. 2) Cumulative-Attention Based Index Selection: This component dynamically selects query-key indexes to be computed based on different attention patterns, ensuring the sum of attention scores meets a predefined threshold. FlexPrefill adaptively optimizes the sparse pattern and sparse ratio of each attention head based on the prompt, enhancing efficiency in long-sequence inference tasks. Experimental results show significant improvements in both speed and accuracy over prior methods, providing a more flexible and efficient solution for LLM inference.
cs.LG · cs.AIJul 7, 2026new
Shervin Khalafi, Igor Krawczuk, Sergio Rozada et al.
Denoising graphs is a fundamental problem in graph learning and the core operation of graph diffusion models. Attention-based architectures like graph transformers have recently shown promise in denoising graphs. However, our principled understanding of attention-based graph denoising remains limited, making it unclear whether standard attention is the right mechanism for this task. Here we show that, under a denoising objective, linear attention is suboptimal and can only learn an average spectral denoising filter over the training distribution. This creates a fundamental limitation as graphs often vary spectrally across the distribution. To overcome this limitation, we introduce Spectral Attention, which directly utilizes the input graph spectrum and provably outperforms linear attention by a margin governed by the spectral diversity of the distribution. We then derive Graph Convolutional Attention (GCA), a practical and permutation-equivariant realization of this idea that implements spectral denoising through graph-filtered queries and keys. For stochastic block models, GCA provably matches the idealized Spectral Attention mechanism. We further show that the softmax operation, that follows the attention, provides additional denoising by approximately projecting noisy eigenvectors onto the clean eigenspace. Empirically, replacing linear attention with GCA consistently improves graph denoising and diffusion on synthetic and real datasets, with gains strongly correlated with spectral diversity. In DiGress, GCA matches standard graph-transformer performance without computing expensive structural features, and when combined with the recently proposed PEARL positional encodings, avoids explicit eigendecomposition computations resulting in faster inference without degrading quality. The code can be found here: github.com/shervinkhalafi/graph_conv_att
paperICLR 20252025143 citations
Seil Kang, Jinyeong Kim, Junhyeok Kim et al.
Large multimodal models (LMMs)"see"images by leveraging the attention mechanism between text and visual tokens in the transformer decoder. Ideally, these models should focus on key visual information relevant to the text token. However, recent findings indicate that LMMs have an extraordinary tendency to consistently allocate high attention weights to specific visual tokens, even when these tokens are irrelevant to the corresponding text. In this study, we investigate the property behind the appearance of these irrelevant visual tokens and examine their characteristics. Our findings show that this behavior arises due to the massive activation of certain hidden state dimensions, which resembles the attention sink found in language models. Hence, we refer to this phenomenon as the visual attention sink. In particular, our analysis reveals that removing the irrelevant visual sink tokens does not impact model performance, despite receiving high attention weights. Consequently, we recycle the attention to these tokens as surplus resources, redistributing the attention budget to enhance focus on the image. To achieve this, we introduce Visual Attention Redistribution (VAR), a method that redistributes attention in image-centric heads, which we identify as innately focusing on visual information. VAR can be seamlessly applied across different LMMs to improve performance on a wide range of tasks, including general vision-language tasks, visual hallucination tasks, and vision-centric tasks, all without the need for additional training, models, or inference steps. Experimental results demonstrate that VAR enables LMMs to process visual information more effectively by adjusting their internal attention mechanisms, offering a new direction to enhancing the multimodal capabilities of LMMs.
cs.CVOct 6, 2026new
Liao Ma, Jiayi Song, Yunfeng Wu et al.
Diffusion Transformers (DiTs) have achieved strong performance in image and video generation, but the quadratic complexity of full attention makes high-resolution generation computationally expensive. Window attention offers an efficient alternative, yet existing methods face a practical trade-off: partitioned window attention typically achieves computational efficiency consistent with its theoretical complexity. However, isolated windows block cross-window interaction, often introducing visible grid-like artifacts in the generated results. Fine-grained sliding-window attention effectively restores interactions across neighboring windows and improves visual quality. However, its irregular computation patterns create a substantial gap between theoretical and practical speedups and require specialized kernels tailored to each hardware backend. To tackle these challenges, we propose BASA, a backend-agnostic sparse attention, which brings the best of both worlds: visual quality and practical acceleration. Specifically, BASA replaces visual self-attention with shifted local-window attention. By introducing a structured window-shifting scheme across DiT blocks, we allow tokens divided by window boundaries in one layer to communicate in the following layers, thereby achieving global information exchange and eliminating window-induced visual artifacts. Notably, our design introduces no additional irregular operators or customized kernels, making it readily deployable on existing attention backends and closing the gap between theoretical sparsity and practical acceleration. Experiments demonstrate that BASA achieves measured speedups exceeding 90\% of the theoretical estimates on FLUX and delivers a 4.52$\times$ attention speedup on Wan while maintaining competitive generation quality.
paperJun 25, 2026▲ 1
Yuzhen Mao, Michael Y. Li, Emily B. Fox
Sparse attention can reduce the cost of long-context inference, but most variants introduce new architectural components. We introduce Simplified Sparse Attention (SSA), a simpler approach to sparse attention that requires no architectural changes. Concretely, we first perform continued pretraining on sequences interleaved with gist tokens. We optimize the standard next-token loss as usual, but the gist tokens use an attention mask to restrict what parts of the context the language model can attend to; this teaches the model to pack each chunk's important information into the gist tokens. At inference time, SSA scores chunks via attention between the current query and the small set of gist tokens, selectively unfolding the top-k chunks by reintroducing their corresponding raw tokens. Since the query is scored only against the gist tokens, we avoid the memory-bandwidth cost associated with naive scoring against the full KV cache, without requiring the auxiliary KV cache approach used by sparse attention methods. On LongBench, SSA consistently outperforms compression and inference-time sparse-attention baselines under the same compression ratio. More strikingly, in retrieval-augmented generation, SSA can even outperform full attention after continued pretraining by over 5.7 points. We attribute this to the ability of SSA's selective unfolding, which concentrates attention on the query-relevant chunks and effectively filters out noise. SSA further extends to a hierarchical gist-of-gist variant (H-SSA) that achieves log-linear decoding complexity while maintaining or improving accuracy at high compression ratios up to 32x. The code is available at https://github.com/yuzhenmao/simplified-sparse-attention/.
paperJun 5, 2026▲ 1
Prateek Kumar Sikdar
Standard transformers apply self-attention uniformly at every layer and token, regardless of whether the input requires dynamic cross-token interaction. We propose CHIAR-Former (Chiaroscuro Attention), a 4-layer hybrid transformer that routes each token to one of three operators - DCT spectral mixing, RBF kernel mixing, or full self-attention - based on per-token spectral entropy, a theoretically justified complexity signal. Through systematic ablation on WikiText-103, we discover routing collapse: the router consistently rejects RBF in favour of DCT and attention, revealing that spectral mixing and dynamic attention are complementary and sufficient. A purpose-designed DCT+Attention-only variant achieves Val PPL 36.54 on WikiText-103 - a 45% improvement over a full-attention baseline (PPL 66.62) at 62.5% fewer attention FLOPs. We extend evaluation to WikiText-2, IMDB sentiment classification, and synthetic ListOps operations, establishing a clear operating regime: CHIAR-Former excels on large-scale naturalistic text where token diversity supports spectral specialisation, while full attention retains an edge on small datasets and synthetic pattern-matching tasks. These findings - both the wins and the losses - together define when and why spectral routing earns its keep.
paperOct 4, 2026▲ 37
Jiarui Chen, Zeqiang Lai, Jiangshan Wang et al.
Sparse attention is a primary approach to reducing the latency of diffusion transformers in long-sequence generation tasks, such as video and high-resolution 3D asset generation. However, existing methods can degrade generation quality and fidelity at high sparsity levels. Through controlled oracle comparisons, we trace this degradation to three sources: constraints imposed by token grouping, inaccurate interaction selection, and the attention contributions lost when tokens are discarded. Guided by this analysis, we propose Meta-Cached Sparse Attention (MC-Sparse), a training-free framework that selects individual key-value (KV) tokens while organizing similar queries into tile-aligned groups for efficient GPU execution. MC-Sparse caches metadata comprising query groups, KV indices selected using exact attention probabilities, and residuals between dense and sparse attention outputs, and reuses them across subsequent denoising steps. Across video and 3D generation models, MC-Sparse achieves higher fidelity to dense-attention outputs and larger denoising speedups than existing sparse-attention baselines, without visible quality degradation. Relative to dense attention, it delivers a 1.80times denoising speedup on Minimax-H3-Base and a 2.32times speedup on 3D asset generation, both with negligible quality loss.
paperJun 28, 2026▲ 41
Disen Lan, Jianbin Zheng, Yuxi Ren et al.
Hybrid attention models improve long-context efficiency by retaining only a subset of full-attention layers and replacing the remaining layers with linear attention. However, the effectiveness of Transformer-to-hybrid conversion critically depends on which layers preserve full attention. Existing hybrid layer selection methods typically rely on heuristic strategies such as fixed placement patterns or layerwise scoring, implicitly treating layer importance as isolated and overlooking the interdependent layer effect under a global hybrid configuration. In this work, we formulate hybrid layer selection as a budget-constrained subset optimization problem. We further propose FlashMorph (Fast LAyer Selection for Hybrid MORPHing), an effective, efficient and scalable layer selection method for Transformer-to-hybrid conversion. FlashMorph first constructs a morphable model by equipping each full-attention layer with a converted linear-attention branch. It then freezes all model weights and jointly optimizes layerwise gates on synthetic long-context retrieval data, with a linearization regularization that encourages the model to rely on linear attention for efficiency. The learned gates are discretized under a preset full-attention budget to instantiate the hybrid architecture, followed by standard logits distillation and long-context finetuning. Extensive experiments show that FlashMorph discovers more effective hybrid configurations, preserves strong long-context recall and general benchmark performance while substantially reducing layer selection cost compared with existing layer selection methods, demonstrating its effectiveness, efficiency, and scalability.
paperJun 12, 2026▲ 3
Ziqing Qiao, Yinuo Xu, Chaojun Xiao et al.
Modern language models increasingly adopt hybrid architectures that combine full attention with efficient attention modules, such as sliding-window attention (SWA) and recurrent sequence mixers. However, how these efficient modules shape model capabilities remains poorly understood. To address this gap, we conduct a systematic analysis across hybrid architectures from three perspectives: scaling behavior, mechanism analysis, and architecture design. First, from a scaling perspective, we find that efficient-attention design primarily affects how fast long-context capability emerges, while different hybrids eventually converge to comparable long-context performance under sufficient training. Second, mechanistically, we show that long-range retrieval is mainly carried by full attention, whereas efficient attention shapes its optimization trajectory. This explains a counter-intuitive phenomenon we call Large-Window Laziness: larger SWA windows can delay the formation of retrieval heads in full-attention layers. Third, guided by this mechanism, we show that applying NoPE to only the full-attention layers of a small-window SWA hybrid substantially improves long-context performance with negligible impact on short-context performance.
cs.AIJun 4, 2026new
Zhuoming Chen, Xinrui Zhong, Qilong Feng et al.
Sparse attention is becoming increasingly important for serving large language models (LLMs) as generation lengths continue to grow. However, deploying and evaluating new sparse attention algorithms at scale remains highly engineering-intensive, slowing both human researchers and AI agents in exploring the sparse attention design. To address this challenge, we present Vortex, a system that combines a Python-embedded frontend language atop a page-centric tensor abstraction for expressing a broad range of sparse attention algorithms, with an efficient backend tightly integrated into modern LLM serving stacks. Vortex enables rapid prototyping, deployment, and evaluation of sparse attention algorithms, effectively translating their theoretical efficiency gains into real-world throughput improvements. As a result, Vortex substantially accelerates the design and iteration of sparse attention algorithms. First, AI agents use Vortex to automatically generate and refine diverse algorithms, the best reaching up to $3.46\times$ higher throughput than full attention while preserving accuracy. Second, Vortex extends sparse attention to emerging architectures and very large models that are otherwise hard to experiment with, reaching up to $4.7\times$ higher throughput on the MLA-based GLM-4.7-Flash and $1.37\times$ on the 229B-parameter MiniMax-M2.7 on NVIDIA B200 GPUs.
paperOct 6, 2026▲ 20
Xiaoran Liu, Ziwei He, Xipeng Qiu
The architectural design of Large Language Models (LLMs) is shifting from traditional full-attention-only models to hybrid models, which combine different attention modules to improve long-context efficiency and performance in length extrapolation and context extension. To explain why hybrid models work and how to design them better, we propose Mechanics of Long-Context Hybrid Models. As Part 1.1 of this series, we begin with hybrids of full attention and either sliding-window attention (SWA) or gated variants of linear attention (LA), represented by GLA and GDN. We first observe a Seesaw Effect in Context Extension: LA hybrids benefit more from long-context continual pretraining, whereas SWA hybrids perform better under length extrapolation. We attribute this behavior to differences in the positional inductive biases induced by these attention mechanisms. We find that SWA hybrids suffer from a Short-Context Learning Trap, Short-Window Weariness, and Long-Window Laziness, and require extended windows to enhance performance in continual long-context pretraining. For LA hybrids, we summarize the Matthew Effect of Hybrid Position Extrapolation and propose Sliding-Window Linear Attention, achieving 16times training-free length extrapolation while maintaining 100\% accuracy on NIAH-SK1 in 64k context length.
paperOct 3, 2026▲ 4
Shuyuan Tu, Qi Tian, Yinming Huang et al.
Natively training joint video-audio generation models at higher resolutions empowers them to learn richer visual details and sharper motion dynamics. However, full attention incurs quadratic cost and, as resolution increases, spreads attention over increasingly redundant tokens, diluting learning signals for informative content and disrupting pretrained priors. Existing sparse attention methods either target training-free acceleration or overlook the unique structure of joint video-audio data, where cross-modal interactions are inherently concentrated around sound-producing regions. To address this, we propose Prism, a dynamic sparse attention framework for natively training joint video-audio generation models at 2K. In particular, Prism organizes the token sequence into spatiotemporal macro-zones, enabling the attention structure to adapt to local content. For each zone, it estimates local information structure via video feature variance along the channel and feature norms from the audio-to-video cross-attention, jointly capturing how visual content varies directionally and how strongly audio influences each visual region. Based on these signals, Prism dynamically assigns a tailored block shape to each zone, applying finer partitioning along axes of rapid visual content variation and strong audio-visual coupling. This encourages tokens within each block to remain semantically coherent, allowing block-level features to capture both visual content and joint video-audio interaction patterns. Prism further adopts a hybrid block selection strategy to dynamically determine per-query sparsity. Experiments show that Prism achieves 2.5times training speedup compared to full attention, while surpassing it in generation quality.
cs.CL · cs.AISep 29, 2026new
Cutter Dawes, Nick Alonso, Tom Figliolia et al.
The attention operation is naively position invariant. However, positional information is fundamental to natural language, and therefore a variety of explicit position encodings have been developed in transformer-based models, such as rotary position encoding (RoPE). Although explicit position encodings have long been assumed to be required, recent methods that interleave local mixing layers, such as sliding window attention (SWA) and gated linear attention, while not encoding position (NoPE) in global attention layers has recently been shown to be successful at scale. How and why this approach works is not well-understood. In this paper, we develop an explanation of how hybrid models of this sort can implicitly encode position at global NoPE layers. Supported by both theoretical and empirical evidence, our central argument is that SWA and gated linear attention induce a recency bias in the residual stream that propagates to, and is selected by, the global attention logits. Moreover, in contrast to the implicit position encodings found in models with only global NoPE attention, in which positional information arises solely from the causal mask, the recency bias in hybrid models can be maintained across long sequences. In addition to deepening our understanding of how hybrid models encode position, these findings may provide insights for how to encode position in a way that can extrapolate to longer sequence lengths indefinitely.
paperAug 27, 2026▲ 31
Jiatong Ding, Bingxin Xing, Yu Zhang et al.
Quantization has been widely adopted in LLM training and inference to reduce cost and improve efficiency. However, low-bit quantization of the attention module often introduces large errors at very low bit-widths, causing performance degradation. Existing methods mainly rely on smoothing techniques to handle outliers, while we propose a hybrid quantization design to better balance accuracy and efficiency. Specifically, we propose HyQuant, an efficient hybrid quantization framework for LLM attention. HyQuant quantizes most attention states into low-bit formats while retaining a small set of vertical-line tokens and local-window states in high precision. These accuracy-critical regions are selected using lightweight vertical-line-aware attention-pattern signals, reducing quantization error with limited overhead. In the Prefill stage, HyQuant uses a hybrid-precision quantized attention operator that preserves vertical-line tokens and a local sliding window in full precision while quantizing the remaining context. In the Decode stage, HyQuant applies the same principle to KV-cache compression and fuses KV dequantization with attention computation to improve memory and hardware efficiency. Across diverse tasks, models, and datasets, HyQuant maintains nearly lossless accuracy with an extremely simple design, demonstrating the efficiency and practical feasibility of hybrid quantization for LLM attention. Code is available at: https://github.com/jerrysfls/HyQuant .
paperAug 27, 2026▲ 14
Alexia Jolicoeur-Martineau, Rhea Sanjay Sukthanker, Pashmina Cameron et al.
Due to the nature of quadratic attention, Large Language Models (LLMs) consume a lot of memory and energy. Every new token costs more than the previous one. For each additional token, the keys and values must be stored in memory indefinitely, which is unsustainable.
Several alternatives have been proposed to fix the quadratic scaling problem, one of which is retrofitting LLMs to use Linear Attention. This idea has attracted a lot of attention, given its promise to solve the quadratic scaling problem with state-of-the-art performance at low cost. However, this line of research has not been properly compared to simpler baselines.
In this work, we show that Sliding Window Attention (SWA) with sinks performs as well or better than post-trained Linear Attention models. We observe this across multiple LLMs on various downstream tasks. For long-context reasoning tasks (Needle-in-a-Haystack and BABILong), SWA achieves massively higher performance (2 to 10 times higher than linear attention). SWA requires no post-training, is extremely fast, and requires low memory; therefore, making it an extremely cheap and reliable solution.
To reduce inference memory cost, we strongly recommend switching to SWA instead of post-training linear models. Linear attention models may have shown some promise, but they likely require to be trained from scratch or extensive post-training in order to even match SWA.
paperJun 10, 2026▲ 126
Xunhao Lai, Weiqi Xu, Yufeng Yang et al.
Ultra-long-context capability is becoming indispensable for frontier LLMs: agentic workflows, repository-scale code reasoning, and persistent memory all require the model to jointly attend over hundreds of thousands to millions of tokens, yet the quadratic cost of softmax attention makes this untenable at deployment scale. We introduce MiniMax Sparse Attention (MSA), a blockwise sparse attention built upon Grouped Query Attention (GQA). A lightweight Index Branch scores key-value blocks and independently selects a Top-k subset for each GQA group, enabling group-specific sparse retrieval while maintaining efficient block-level execution; the Main Branch then performs exact block-sparse attention over only the selected blocks. Designed around a principle of simplicity and scalability, MSA is deliberately streamlined, making it straightforward to deploy efficiently across a broad range of GPUs. To translate sparsity into practical speedups, we co-design MSA with a GPU execution path that uses exp-free Top-k selection and KV-outer sparse attention to improve tensor-core utilization under block-granular access. On a 109B-parameter model with native multimodal training, MSA performs on par with GQA while reducing per-token attention compute by 28.4x at 1M context. Paired with our co-designed kernel, MSA achieves 14.2x prefill and 7.6x decoding wall-clock speedups on H800. Our inference kernel is available at: https://github.com/MiniMax-AI/MSA. A production-grade natively multimodal model powered by MSA has been publicly released at: https://huggingface.co/MiniMaxAI/MiniMax-M3.
paperNeurIPS 20242024287 citations
Dongchen Han, Ziyi Wang, Zhuofan Xia et al.
Mamba is an effective state space model with linear computation complexity. It has recently shown impressive efficiency in dealing with high-resolution inputs across various vision tasks. In this paper, we reveal that the powerful Mamba model shares surprising similarities with linear attention Transformer, which typically underperform conventional Transformer in practice. By exploring the similarities and disparities between the effective Mamba and subpar linear attention Transformer, we provide comprehensive analyses to demystify the key factors behind Mamba's success. Specifically, we reformulate the selective state space model and linear attention within a unified formulation, rephrasing Mamba as a variant of linear attention Transformer with six major distinctions: input gate, forget gate, shortcut, no attention normalization, single-head, and modified block design. For each design, we meticulously analyze its pros and cons, and empirically evaluate its impact on model performance in vision tasks. Interestingly, the results highlight the forget gate and block design as the core contributors to Mamba's success, while the other four designs are less crucial. Based on these findings, we propose a Mamba-Inspired Linear Attention (MILA) model by incorporating the merits of these two key designs into linear attention. The resulting model outperforms various vision Mamba models in both image classification and high-resolution dense prediction tasks, while enjoying parallelizable computation and fast inference speed. Code is available at https://github.com/LeapLabTHU/MLLA.
cs.CVOct 13, 2026new
Junfeng Ni, Ziyu Zhu, Fan Jiang et al.
Video diffusion models generate visually compelling short-term sequences but often struggle to preserve scene geometry and object structure over time. In this paper, we first demonstrate that current video attention cannot identify reliable correspondences across frames, with responses dispersed across multiple regions even after recognizable scene structure has emerged. We argue that reliable cross-frame associations are essential for information exchange as scene structure forms in video generation. Thus, we introduce Geometry Attention Distillation (GAD), which transfers correspondences from a frozen geometry model's attention to video attention. Extracted from clean videos, these correspondences link the same static surface or moving object across frames. Since structure formation begins in deeper layers at higher noise and shallower layers at lower noise, we apply attention distillation at a student layer selected according to the noise timestep. After training, video attention focuses more consistently on corresponding locations. Experiments across all three video backbones demonstrate improved long-range scene consistency and dynamic temporal consistency, with geometric gains across generation lengths. On Wan2.2 5B, GAD improves cross-frame matching accuracy by at least 30% over the Wan baseline on both long-range benchmarks. Notably, GAD achieves these gains without additional inference modules, offering a promising approach to efficient and consistent video-based world simulation. Project page: https://fantasy-amap.github.io/GAD/.
paperOct 11, 2026▲ 5
Javad Rajabi, Koorosh Roohi, David B. Lindell et al.
Generating images beyond the training resolution of Diffusion Transformers (DiTs) remains challenging, as the number of tokens grows rapidly with resolution, full attention becomes expensive, and outputs often suffer from structural distortions and degraded detail. We identify uniform full attention as the common source of both problems. Attending to all tokens at every step causes quadratic cost and exposes the model to more tokens and relative distances beyond the training range, pushing the model out of distribution. We observe that attention's effective receptive field narrows during denoising, from global in early steps that form structure to local in later steps that refine details. We propose AReA, a training-free attention mechanism whose receptive field evolves from coarse to fine, following the denoising progression. In early steps, AReA applies cross-scale coarse attention over downsampled keys and values, capturing global context, while in later steps it applies local attention within progressively shrinking windows. Without additional parameters or training, experiments show that AReA consistently outperforms state-of-the-art training-free baselines across target resolutions and different models, while also being faster and more efficient.
paperOct 7, 2026▲ 5
Avihai Giuili, Rotem Atari, Avishai Sintov et al.
Learned simulators for deformable linear objects (DLOs) such as cables have to predict the motion of cables they were not trained on and stay stable over long rollouts. Most of their error occurs where the cable touches itself or the floor. Attention over all pairs of cable segments can represent contact between parts of the cable that are far apart along its length, but attention has no notion of geometry. A cable has two pairwise distances, which agree only while it is straight: the arc-length distance along the cable, which governs elastic forces, and the Euclidean distance in space, which governs contact. We add a physical attention bias, an additive term on the attention logits with a learned rate, and ask which distance it should use. We compare no bias, each distance alone, and both distances on disjoint sets of heads, keeping the rest of the model and the training protocol fixed. A physical bias improves prediction on unseen cables. The gain is largest when attention is the only mechanism that connects distant segments: there, the arc-length bias reduces prediction error by 15% and more than halves the drift in segment length. The Euclidean bias alone stays close to unbiased attention, while assigning both distances across heads is best or near-best on every metric we report. Code and per-run records: https://github.com/avihaig/dlogps.
paperSep 28, 2026▲ 3
Oliver Sieberling, Bharat Runwal, David Jin et al.
Recurrent neural networks (RNNs) compress the historical context into a memory state of fixed size, thus allowing for constant-time inference. The memory state size is a crucial factor in their performance, as exemplified by the strong performance and resurgence of linear attention, which extends the vector-valued hidden states of ordinary RNNs to matrix-valued hidden states. Crucially, linear attention does so in a parameter-efficient way, in particular by using an outer product of the key and value vectors to write to the matrix-valued hidden state. We generalize this construction and propose triadic linear attention, which writes the triadic outer product of a key, a second key, and a value, into a third-order (i.e., 3D) tensor state, and reads from it by contracting both key axes with two queries. An E-dimensional second key thus yields an E-fold increase in state size while adding only two projections. Triadic linear attention is compatible with data-dependent forgetting, the delta rule, and chunkwise-parallel training. Applied to Gated DeltaNet and scalar-gated linear attention, triadic linear attention substantially improves long-context language modeling and recall, outperforming alternatives that enlarge the state.
cs.CLSep 17, 2026new
Haibo Feng, Ruiqi Liang, Hanyang Peng et al.
Reasoning and agentic workloads increasingly demand efficient long-context inference. Yet full-attention decoding reads the growing history at every step, regardless of its benefit to the next prediction. We show that a pretrained model's decoding states already contain information predictive of this benefit, before the global read. Building on this finding, we introduce On-Demand Attention (ODA), a local-first decoding method that uses a lightweight recall head to selectively invoke global attention as its predicted benefit changes during generation. ODA trains only the recall head, leaving pretrained weights unchanged and the complete historical KV cache available for future recall. We further implement GPU-side conditional execution in vLLM, translating reduced global reads into practical decoding speedups over full attention at long context lengths. Experiments across Qwen and Gemma models, including hybrid-attention backbones, show that selective recall recovers most of the performance lost under local attention while substantially reducing global reads. These findings support long-context inference in which pretrained models guide their own access to the information they retain.
paperJul 26, 2026▲ 28
Haopeng Li, Yitong Li, Junsong Chen et al.
Diffusion transformers are essential for high-fidelity video generation, but long token sequences make attention a dominant inference bottleneck. Training-free dynamic sparse attention alleviates this bottleneck by computing only selected key-value blocks, yet existing methods struggle to sparsify attention both efficiently and accurately for two reasons: (1) Rigid, unpredictable, and costly routing: selecting a fixed fraction of top-ranked blocks by proxy score imposes fixed budgets, whereas retaining blocks to reach a target cumulative proxy probability mass yields dynamic but potentially imbalanced budgets; both incur non-negligible overhead from computing and materializing proxy scores. (2) Lossy keep-or-drop sparsification: unselected blocks are discarded entirely, degrading accuracy under aggressive sparsity. These limitations motivate cheaper dynamic-budget routing while limiting accuracy degradation. In this paper, we introduce training-free Sol-Attn (Sparsifying online attention), which unifies dynamic routing, sparse computation, and approximation correction in a single online-softmax pass, achieving a better accuracy-efficiency trade-off in sparse attention. The core of Sol-Attn is on-the-fly block thresholding with proxy-score reuse, which selects critical blocks by comparing block proxy scores against a threshold during online softmax. This design enables dynamic yet controllable block budgets without materializing the proxy map, while directly reusing the proxy scores of unselected blocks to approximate their contribution. Experiments across image and video generation tasks show that Sol-Attn advances the quality-efficiency frontier of training-free sparse attention, delivering 2.1 times and 2.3 times end-to-end speedups for video generation and editing, respectively, while preserving visual quality.
paperJul 16, 2026▲ 8
Hao Liu, Chenghuan Huang, Ye Huang et al.
Video Diffusion Transformers process long spatio-temporal sequences, making self-attention the main bottleneck in high-resolution video generation. Training-free sparse attention reduces this cost, but adaptive Top-p routing creates uneven per-head workloads under multi-GPU sequence parallelism. The resulting workload heterogeneity turns sparse attention into a rank-level straggler problem. We present , a training-free sparse-attention system that improves the distributed execution efficiency of adaptive sparse attention under multi-GPU sequence parallelism. uses Top-p routing, a Top-k safety floor, and video-aware block organization as the sparse-routing frontend, then repairs the materialized mask at runtime. Runtime Load Balancing migrates a small number of heavy heads via P2P communication to shorten the current critical path. Slack-Aware Sparse Augmentation fills residual non-critical-rank slack with additional high-value blocks, while overlap hides scheduling and migration overhead behind existing computation. On step-distilled Wan2.2 I2V, reduces average load imbalance from 1.34 to 1.08 and delivers a 4.41times attention speedup over FlashAttention, while achieving a 2.02--2.11times DiT inference speedup with competitive video quality.
cs.CVJun 29, 2026new
Seongro Yoon, Donghyeon Cho, Jinsun Park et al.
Understanding facial expressions in videos requires modeling subtle and localized facial dynamics under unconstrained conditions. Although recent Vision Transformer~(ViT)-based video models have shown strong performance through large-scale self-supervised pretraining, their attention mechanisms often emphasize dominant global motions and coarse temporal dynamics, limiting sensitivity to fine-grained facial variations. To address this limitation, we propose MiRA (Marginal-induced Attention Redistribution), a plug-in frame-marginal attention redistribution framework for ViT backbones that enhances spatio-temporal selectivity toward subtle facial dynamics without introducing additional trainable parameters. MiRA derives frame-level confidence and intra-frame concentration statistics from self-attention maps to estimate frame-wise marginal importance and redistribute attention toward spatiotemporally localized facial cues. We first introduce a principled \textit{exact mode} based on post-softmax attention redistribution. To further improve efficiency, we propose \textit{flashLite mode}, a lightweight pre-softmax approximation that integrates frame-marginal redistribution into FlashAttention kernels while preserving the effectiveness of the exact formulation. Experimental results on challenging Facial Expression Recognition~(FER) benchmarks demonstrate consistent improvements over strong ViT baselines.
paperJun 17, 2026▲ 13
Zhentao Tan, Wei Chen, Jingyi Shen et al.
The quadratic complexity of attention poses a critical bottleneck for long-context processing, spurring interest in hybrid attention designs. Most open-source hybrid models adopt a layer-wise strategy. Yet, prior work has noted the inherent difficulty of integrating Linear Attention (LA) with Full Attention (FA), suggesting that the design space of attention hybridization remains underexplored. To probe this space, we conduct interpretability analysis and observe that layers exhibit block-wise functional similarity, while individual heads within the same layer display distinct functional specialization despite sharing input features. This head-level heterogeneity suggests that the head dimension provides a natural and principled granularity for fusing heterogeneous attention signals. Building on this insight, we introduce HydraHead, a novel architecture that hybridizes FA and LA along the head axis. HydraHead features two key innovations: (1) an interpretability-driven selection strategy that identifies retrieval-critical heads and preserves FA only for them, and (2) a scale-normalized fusion module that reconciles the distributional gap between FA and LA head outputs. By leveraging a three-stage transfer pipeline with parameter reuse and distillation, we achieve high-performance hybrid models with minimal training overhead. Under a unified training setup, HydraHead outperforms other hybrid designs in long-context tasks while maintaining strong general reasoning. With interpretability-driven head selection, it matches a 3:1 layer-wise hybrid's long-context performance at a 7:1 LA-to-FA ratio. Crucially, trained on only 15B tokens, HydraHead achieves over 69% improvement over the baseline at 512K context length, approaching Qwen3.5, a leading model of comparable size with a native context length of 256K. This highlights the significant scaling potential of head-level hybridization.