HSMLA: Hierarchical Softmax Multi-scale Linear Attention for Efficient Vision Transformers

Vision transformers face significant computational overheads in high-resolution dense prediction due to the quadratic complexity of self-attention. Linear attention offers efficiency but sacrifices local context modeling. We propose \textbf{HSMLA (Hierarchical Softmax Multi-scale Linear Attention)}, which combines ReLU-based linear attention for global context, selective softmax refinement for critical local features, and multi-scale token representations via depthwise convolutions. HSMLA achieves superior accuracy-efficiency trade-offs: up to $4.2\times$ inference-time speedup across dense prediction tasks, $87.3%$ Dice with $3.2\times$ speedup on CT organ segmentation, and $94.2%$ AUC with $4.1\times$ speedup on pathology WSI.

RUBRIC: Realism--Utility Balanced Ranking for Imbalanced Classification

Class imbalance poses a fundamental challenge in risk-sensitive applications such as fraud detection and medical diagnosis, where minority-class samples are scarce yet critical for accurate classification. Existing oversampling methods generate synthetic samples to rebalance class distributions; however, they often produce large numbers of low-quality candidates that distort decision boundaries or introduce artifacts, leading to overfitting and degraded generalization. In this work, we introduce RUBRIC, a generator-agnostic filtering framework that formulates synthetic sample selection as a quality-over-quantity optimization problem. RUBRIC ranks candidates using a realism-utility trade-off: realism is quantified by a learned discriminator that distinguishes real samples from synthetic samples, while utility captures proximity to the decision boundary through a concave margin-based scoring function. We show that, under mild regularity conditions, the proposed filtering strategy monotonically tightens the generalization bound for margin-based classifiers by jointly reducing distribution shift and suppressing near-negative tail contributions. Through extensive experiments on credit-card fraud detection and other imbalanced benchmarks, we demonstrate that RUBRIC improves F1-macro and recall while maintaining comparable ROC-AUC across several generators. We also provide explicit lambda-sensitivity analysis to show how users can recover AUPRC when ranking quality is prioritized.

OLIVE: Online Low-Rank Incremental Learning for Efficient Adaptive Exoskeletons

Wearable exoskeleton systems hold promise for restoring mobility in individuals with physical impairments, yet most existing controllers rely on static gait policies that lack the ability to adapt to dynamic real-world environments or individual user characteristics. We present \olive (\underline{O}nline \underline{L}ow-rank \underline{I}ncremental Learning for Efficient Adapti\underline{ve} Exoskeletons), a parameter-efficient online adaptation framework that continuously personalizes exoskeleton control during deployment. \olive decomposes the adaptive component of the control policy into a low-rank residual form~$\dW = \At\Bt^\top$ with rank~$r!\ll!\min(d,k)$, reducing online update cost from $\mathcal{O}(dk)$ to $\mathcal{O}(r(d{+}k))$ while preserving the stability of a pretrained base controller~$\Wz$. Parameters are updated via a reward-shaped policy gradient driven purely by on-body sensor feedback (EMG, IMU, vibration), eliminating dependence on offline reference trajectories. A gating mechanism modulates the strength of personalization based on contextual state, and a dynamic rank scheduler adapts the update dimensionality to terrain complexity -- allocating minimal capacity on simple flat terrain and expanding to higher-rank updates on demanding uneven surfaces -- enabling robust performance across diverse activities: flat walking, stair navigation, slopes, and uneven terrain. Experiments on the wearable platform demonstrate that \olive achieves +13, +22, and +15 percentage-point improvements in gait smoothness, effort reduction, and motion stability over the strongest baseline, converging within $\sim$1{,}800 walking steps at 7.4,ms end-to-end latency. Our code implementation is available at https://github.com/FastLM/OLIVE.

Accelerating Frequency Domain Diffusion Models with Error-Feedback Event-Driven Caching

Diffusion models achieve remarkable success in time series generation. However, slow inference limits their practical deployment. We propose E$^2$-CRF (Error-Feedback Event-Driven Cumulative Residual Feature caching) to accelerate frequency domain diffusion models. Our method exploits two structural properties: (1) spectral localization, where signal energy concentrates in low frequencies, and (2) mirror symmetry, which halves the effective frequency dimension. E$^2$-CRF uses a closed-loop error-feedback system that adaptively caches transformer KV features across diffusion steps. We trigger recomputation using event-driven residual dynamics instead of fixed schedules. Our method selectively recomputes high-energy or rapidly-changing tokens while reusing cached features for stable high-frequency components. E$^2$-CRF achieves ~2.2 speedup while maintaining sample quality. We demonstrate effectiveness on 5 datasets. Our caching strategy naturally aligns with the diffusion process's structure-to-detail progression. We include sufficient-condition error and complexity bounds under standard regularity assumptions (Appendix), alongside empirical validation. Our code is available at https://github.com/NoakLiu/FastFourierDiffusion and is also integrated in https://github.com/NoakLiu/FastCache-xDiT.

MKA: Memory-Keyed Attention for Efficient Long-Context Reasoning

As long-context language modeling becomes increasingly important, the cost of maintaining and attending to large Key/Value (KV) caches grows rapidly, becoming a major bottleneck in both training and inference. While prior works such as Multi-Query Attention (MQA) and Multi-Latent Attention (MLA) reduce memory by sharing or compressing KV features, they often trade off representation quality or incur runtime overhead. We propose Memory-Keyed Attention (MKA), a hierarchical attention mechanism that integrates multi-level KV caches (local, session, and long-term) and learns to route attention across them dynamically. We further introduce Route-Fused MKA (FastMKA), a broadcast-routed variant that fuses memory sources before attention computation for improved efficiency. Experiments on different sequence lengths show that FastMKA achieves a favorable accuracy-efficiency trade-off: comparable perplexity to MLA while achieving up to 5x faster training throughput and 1.8x lower evaluation latency. These results highlight MKA as a practical and extensible framework for efficient long-context attention.

AdaCorrection: Adaptive Offset Cache Correction for Accurate Diffusion Transformers

Diffusion Transformers (DiTs) achieve state-of-the-art performance in high-fidelity image and video generation but suffer from expensive inference due to their iterative denoising structure. While prior methods accelerate sampling by caching intermediate features, they rely on static reuse schedules or coarse-grained heuristics, which often lead to temporal drift and cache misalignment that significantly degrade generation quality. We introduce \textbf{AdaCorrection}, an adaptive offset cache correction framework that maintains high generation fidelity while enabling efficient cache reuse across Transformer layers during diffusion inference. At each timestep, AdaCorrection estimates cache validity with lightweight spatio-temporal signals and adaptively blends cached and fresh activations. This correction is computed on-the-fly without additional supervision or retraining. Our approach achieves strong generation quality with minimal computational overhead, maintaining near-original FID while providing moderate acceleration. Experiments on image and video diffusion benchmarks show that AdaCorrection consistently improves generation performance.

CXL-SpecKV: A Disaggregated FPGA Speculative KV-Cache for Datacenter LLM Serving

Large Language Models (LLMs) have revolutionized natural language processing tasks, but their deployment in datacenter environments faces significant challenges due to the massive memory requirements of key-value (KV) caches. During the autoregressive decoding process, KV caches consume substantial GPU memory, limiting batch sizes and overall system throughput. To address these challenges, we propose \textbf{CXL-SpecKV}, a novel disaggregated KV-cache architecture that leverages Compute Express Link (CXL) interconnects and FPGA accelerators to enable efficient speculative execution and memory disaggregation. Our approach introduces three key innovations: (i) a CXL-based memory disaggregation framework that offloads KV-caches to remote FPGA memory with low latency, (ii) a speculative KV-cache prefetching mechanism that predicts and preloads future tokens' cache entries, and (iii) an FPGA-accelerated KV-cache compression and decompression engine that reduces memory bandwidth requirements by up to 4$\times$. When evaluated on state-of-the-art LLM models, CXL-SpecKV achieves up to 3.2$\times$ higher throughput compared to GPU-only baselines, while reducing memory costs by 2.8$\times$ and maintaining accuracy. Our system demonstrates that intelligent memory disaggregation combined with speculative execution can effectively address the memory wall challenge in large-scale LLM serving. Our code implementation has been open-sourced at https://github.com/FastLM/CXL-SpecKV.

$π$-Attention: Periodic Sparse Transformers for Efficient Long-Context Modeling

Transformers have revolutionized natural language processing, but their quadratic complexity with respect to sequence length remains a fundamental bottleneck for long-range modeling. While sparse attention mechanisms like RingAttention reduce computational costs by restricting attention to local neighborhoods, they suffer from limited receptive fields and lack of adaptability. We present \PiAttention, a periodic sparse Transformer that factorizes attention into ring-local neighborhoods, deterministic $π$-stride skips, and an adaptive fusion gate. The periodic structure provides predictable coverage of distant tokens, while the sparse footprint keeps the per-layer complexity linear in context length. We prove that \PiAttention achieves $\mathcal{O}(kL + π\log L)$ receptive field growth compared to $\mathcal{O}(kL)$ for RingAttention, where $k$ is the local window size, $π$ is the skip period, and $L$ is the sequence length. Extensive experiments on language modeling, retrieval, and vision-language tasks demonstrate that \PiAttention matches or surpasses dense attention quality with 8.3\% lower perplexity than RingAttention while using 50\% fewer GPUs for the same context length. Our detailed ablations and visualizations reveal the importance of periodic skips, adaptive fusion, and head-level sparsity coordination for efficient long-context modeling.

SemToken: Semantic-Aware Tokenization for Efficient Long-Context Language Modeling

Tokenization plays a critical role in language modeling, yet existing approaches such as Byte-Pair Encoding (BPE) or WordPiece operate purely on frequency statistics, ignoring the underlying semantic structure of text. This leads to over-tokenization of semantically redundant spans and underutilization of contextual coherence, particularly in long-context scenarios. In this work, we propose \textbf{SemToken}, a semantic-aware tokenization framework that jointly reduces token redundancy and improves computation efficiency. SemToken first extracts contextual semantic embeddings via lightweight encoders and performs local semantic clustering to merge semantically equivalent tokens. Then, it allocates heterogeneous token granularity based on semantic density, allowing finer-grained tokenization in content-rich regions and coarser compression in repetitive or low-entropy spans. SemToken can be seamlessly integrated with modern language models and attention acceleration methods. Experiments on long-context language modeling benchmarks such as WikiText-103 and LongBench show that SemToken achieves up to $2.4\times$ reduction in token count and $1.9\times$ speedup, with negligible or no degradation in perplexity and downstream accuracy. Our findings suggest that semantic structure offers a promising new axis for optimizing tokenization and computation in large language models.

FastCache: Fast Caching for Diffusion Transformer Through Learnable Linear Approximation

Diffusion Transformers (DiT) are powerful generative models but remain computationally intensive due to their iterative structure and deep transformer stacks. To alleviate this inefficiency, we propose FastCache, a hidden-state-level caching and compression framework that accelerates DiT inference by exploiting redundancy within the model's internal representations. FastCache introduces a dual strategy: (1) a spatial-aware token selection mechanism that adaptively filters redundant tokens based on hidden state saliency, and (2) a transformer-level cache that reuses latent activations across timesteps when changes are statistically insignificant. These modules work jointly to reduce unnecessary computation while preserving generation fidelity through learnable linear approximation. Theoretical analysis shows that FastCache maintains bounded approximation error under a hypothesis-testing-based decision rule. Empirical evaluations across multiple DiT variants demonstrate substantial reductions in latency and memory usage, with best generation output quality compared to other cache methods, as measured by FID and t-FID. Code implementation of FastCache is available on GitHub at https://github.com/NoakLiu/FastCache-xDiT.