每日文章

Data-Free Quantization (DFQ) addresses data security concerns by synthesizing samples, without accessing real data. It has garnered increasing attention in the context of Vision Transformers (ViTs), owing to the superiority of the self-attention mechanism compared to classical convolutional operation. However, previous DFQ arts for ViTs often suffer from a distribution mismatch between synthetic samples and input distribution expected by quantized models Q, resulting in the suboptimal performance. In this paper, we propose a novel Masked Attention Alignment approach for Data-Free Quantization of ViTs, named MaskAQ, revealing that: 1) the semantics in the self-attention mechanism is predominantly localized to a sparse subset of patches, called informative regions; 2) the informative regions dominate the mutual information between synthetic samples and Q's outputs. To these ends, we incorporate differential entropy maximum over patch similarity of synthetic samples, to decouple informative regions from noisy background. To couple with varied Q, the informative regions are selected to align full-precision models with Q via a masked attention alignment objective, thus yielding high-quality synthetic samples. Furthermore, a periodic sample refreshing strategy comes up to endow MaskAQ with the capacity to continually adapt to the evolving state of Q throughout the training process, to preserve desirable mutual information with synthetic samples. Extensive experiments verify the merits of MaskAQ over state-of-the-art approaches across multiple backbones and downstream tasks. Our code is available at https://github.com/hfutqian/MaskAQ.

Mixture-of-Experts (MoE) architectures scale model capacity through sparse expert activation, but their deployment remains memory-bound because all expert weights must reside in memory. Mixed-precision quantization can substantially reduce this footprint by assigning different bit-widths to different experts. Existing approaches, however, typically rely on calibration data to estimate expert importance and determine bit allocation. For frontier MoE LLMs, the original training data, and hence the true training distribution, is proprietary and inaccessible. As a result, calibration sets are inevitably imperfect surrogates, and this can misestimate expert utilization and lead to suboptimal bit allocation. Motivated by the substantial cross-expert quality variability observed in modern MoE models, and by the success of Heavy-Tailed Self-Regularization (HT-SR) theory at predicting neural network model quality without access to training or testing data, we propose AlphaQ, a calibration-free bit-allocation method for MoE quantization. AlphaQ draws on HT-SR theory and follows a simple principle: experts with more heavy-tailed weight spectra are typically better trained and hence should receive higher bit-widths, while experts with weaker heavy-tailed structure can be quantized more aggressively. AlphaQ operationalizes this principle by measuring expert-wise spectral heavy-tailedness and solving a budget-constrained optimization problem that minimizes total quantization error under a global bit-budget constraint. Across several MoE models, AlphaQ consistently outperforms calibration-based baselines under matched bit budgets. Notably, on Qwen1.5-MoE, AlphaQ achieves near full-precision accuracy with an average expert precision of only 3.5 bits, while delivering more than 4\(\times\) memory compression. Our code is available at https://github.com/Superone77/AlphaQ.

Quantizing deep neural networks is essential for efficient inference on resource-constrained devices. However, most existing methods are designed for single-domain and class-balanced data, leaving practical settings with domain shifts or severe class imbalance underexplored. We address these challenges with Efficient Multi-Domain Alignment Quantization (EmaQ), which aligns domain distributions through a CDF-based projection and uses sensitivity-aware weight aggregation to stabilize multi-domain quantization. We further extend EmaQ to EmaQ-LT for long-tailed quantization by introducing class-conditioned variance scaling and confidence-based logit adjustment to mitigate majority-class overconfidence. Theoretical analyses establish convergence guarantees and motivate the proposed sensitivity and scaling mechanisms. Experiments on standard, multi-domain (Office-31, Digits), and long-tailed (SynDigits-LT, CIFAR-10-LT, CIFAR-100-LT) benchmarks show that EmaQ and EmaQ-LT achieve strong low-bit performance under domain shift and class imbalance.

Diffusion large language models (DLLMs) have recently emerged as a promising alternative to autoregressive LLMs by generating text through iterative masked denoising with bidirectional context. However, their large model sizes and iterative denoising process introduce substantial memory and computational overhead, motivating post-training quantization for efficient deployment. In this paper, we identify two key challenges for low-bit DLLM quantization: state-dependent activation disparity and temporal error accumulation. Masked and unmasked tokens exhibit different activation distributions within each denoising step, while quantization errors can accumulate across steps during iterative decoding. To address these challenges, we propose STaR-Quant, a state-time consistent PTQ framework for DLLMs. STaR-Quant introduces State-Guided Activation Transformation (SGAT) to assign masked and unmasked tokens to different activation transformation spaces with a unified static weight-side transformation. It further introduces Temporal Attention Compensation (TAC) to correct the quantized attention representation via a lightweight block-diagonal affine mapping. Experiments on representative DLLMs demonstrate that STaR-Quant consistently improves low-bit weight-activation quantization over strong PTQ baselines, while delivering up to 1.69x speedup and 3.14x memory saving over FP16 deployment.

Sparse attention reduces compute and memory bandwidth for long-context LLM inference. However, two key challenges remain: (1) KV cache capacity still grows with sequence length, and offloading to CPU memory introduces a PCIe transfer bottleneck; (2) the sparse selection step itself retains \(O(T^2)\) complexity and can dominate attention cost at long contexts. We propose SparDA, a decoupled sparse attention architecture that introduces a fourth per-layer projection, the Forecast, alongside Query, Key, and Value. The Forecast predicts the KV blocks needed by the next layer, enabling lookahead selection that overlaps CPU-to-GPU prefetch with current-layer execution. Because Forecast is decoupled from the attention query, our GQA implementation uses one Forecast head per GQA group, reducing selection overhead versus the original multi-head selector. SparDA adds \(<\)0.5% parameters and trains only the Forecast projections by matching the original selector's attention distribution. On two sparse-pretrained 8B models, SparDA matches or slightly improves accuracy and delivers up to 1.25\(\times\) prefill speedup and 1.7\(\times\) decode speedup over the sparse-attention offload baseline. By enabling larger feasible batch sizes on a single GPU, SparDA further reaches up to 5.3\(\times\) higher decode throughput than the non-offload sparse baseline. Our source code is available at https://github.com/NVlabs/SparDA.

Key-value (KV) caching accelerates inference of large language models (LLMs) by reusing past computations for generated tokens. Its importance becomes even greater in long-context applications such as retrieval-augmented generation (RAG) and in-context learning (ICL). However, conventional KV caching embeds positional information directly into the cache, limiting its reusability. Existing solutions either restrict reuse to prefixes or require expensive memory materialization for positional re-encoding. We introduce LazyAttention, a novel attention mechanism that kernelizes deferred positional encoding to enable zero-copy, position-agnostic KV reuse. By adjusting positional encoding within attention kernels on-the-fly, LazyAttention resolves the materialization bottleneck, allowing a single physical KV copy to serve multiple logical requests at arbitrary positions. Leveraging attention kernels tailored for prefilling and decoding, our system achieves significant efficiency improvements: under skewed document distributions, it reduces time-to-first-token (TTFT) by 1.37\(\times\) and increases inference throughput by 1.40\(\times\) compared to the state-of-the-art Block-Attention, while maintaining comparable output quality.

Video diffusion transformers have achieved state-of-the-art visual quality, but their high inference cost remains a major bottleneck for real-time applications. Recent distillation frameworks produce autoregressive video diffusion models with reduced latency, yet these models still use a fixed number of denoising steps per frame, wasting computation on predictable frames and under-refining challenging ones. We present DSA, a confidence-guided adaptive computation framework for AR video diffusion. DSA introduces a lightweight confidence head, trained jointly with the generator under a distribution-matching distillation objective, to estimate per-frame denoising reliability. At inference, this confidence signal dynamically adjusts the number of diffusion steps: simple frames terminate early for speed, while complex frames receive additional refinement. Our method requires no extra video data, no heuristics, and little architectural modification. Experiments show that DSA achieves real-time autoregressive video generation, reaching 22.63 FPS with sub-second latency on H100 GPUs, while maintaining competitive or superior VBench quality compared to recent autoregressive and bidirectional video diffusion models. Our results demonstrate that confidence-guided adaptive sampling provides an effective and practical path toward interactive video generation.

We present Echo Infinity, an autoregressive (AR) framework towards real-time infinite video generation that employs a learnable evolving memory to dynamically filter, abstract, and compress any-length history at constant cost. Existing methods mainly curate memory with predefined KV-cache schedules, fixed-ratio heuristic compression, or inference-time RoPE adaptation. These designs inevitably lose historical information and amplify compounding errors due to their limited cache window and ignorance of autoregressive generation noise. Inspired by human memory consolidation, Echo-Infinity replaces handcrafted memory curation with learnable Memory Query, which are updated by attention and a gating mechanism when past frames are evicted from the local window. The queries are optimized end-to-end with the video diffusion transformers (DiTs), forming an evolving memory that supports arbitrary compression ratios with constant computation independent of video length. They also act as a generalizable generation prior, improving quality even when only the optimized initial state is used. We further introduce Unified Relative RoPE Recipe, which anchors the sink frames to start from id 0 and lets the newest frame id grow at most to the DiTs' pretrained maximum temporal RoPE id throughout training and inference, freeing the model from the finite RoPE constraint and closing the train-test RoPE extrapolation gap. In long and short video generation, Echo-Infinity achieves state-of-the-art performance, and, to our knowledge, demonstrates promising 24-hour (>1.3 M frames) real-time rollouts for the first time, suggesting a practical path toward infinite video generation.

Spiking Vision Transformer (SViT) models are promising low-power ViT models for solving vision-based tasks with state-of-the-art performance. However, their large sizes limit their deployments for resource-constrained embedded platforms, underscoring the needs of model compression. One of prominent compression techniques is pruning, and the state-of-the-art works employ unstructured pruning techniques to compress SViT models. Such techniques require specialized hardware architectures tailored for the sparsity patterns to maximize their efficiency benefits, making this approach not scalable. To address this, we propose PSViT, a novel methodology to perform structured pruning on SViT models, hence making it possible to efficiently accelerate their inference using the existing and widely-used computing architectures. To do this, PSViT employs several key steps: uniform channel-wise filter pruning to structurally eliminate the non-significant weights, sensitivity analysis to evaluate the impact of channel-wise pruning of individual layer on accuracy and network size, as well as fine-grained channel-wise pruning based on the sensitivity analysis and the given network architecture. Experimental results show that PSViT effectively obtains 22.4% memory saving through single-shot pruning, while maintaining high accuracy within 3% (70.3% without fine-tuning and 72.8% with fine-tuning) from the original non-pruned SViT model (73.3%) on the ImageNet-1K. These results also show that the PSViT methodology advances the effort in enabling efficient SViT deployments on resource-constrained applications.

The large sizes of Spiking Vision Transformers (SViTs) still hinder their embedded implementation, highlighting the need for model compression. State-of-the-art works compress SViT models through unstructured pruning, which needs specialized hardware accelerators for their specific sparsity patterns to maximize efficiency gains. Moreover, their manual approach requires a huge design time to find an appropriate pruning setting for each network, thus making this approach not scalable. To address this limitation, we propose PrimeSVT, a novel framework that performs automated memory-aware structured pruning on pre-trained SViT models, thereby maximizing their efficiency gains during inference amenable to widely-used computing architectures. To achieve this, PrimeSVT first sorts the SViT layers based on their sizes (i.e., number of parameters), identifies the targeted pruning layers based on their robustness under different pruning rates, then leverages this order for compressing the model layer-by-layer sequentially from the largest one to the smallest one (i.e., so-called prioritized compression policy), while considering the user-defined constraints (i.e., acceptable accuracy and memory saving). In each layer, PrimeSVT employs channel-wise filter pruning based on their L2-norm values to structurally remove the non-significant weights. Experimental results show that PrimeSVT saves 26.68% memory through automated single-shot pruning, while preserving accuracy within 3% (70.3% without fine-tuning and 72.9% with fine-tuning) from the original unpruned SViT model (73.3%), thus meeting the accuracy and memory constraints. These show that our PrimeSVT framework enables design automation for SViTs and their embedded implementation.

One-shot block drafters for speculative decoding generate the full draft in a single forward pass, achieving strong throughput by eliminating sequential token generation. However, they predict each draft token conditioned only on the prefix context, with no dependence on previously drafted tokens. This non-autoregressive conditioning causes the drafter's distribution to diverge from the verifier's true autoregressive distribution as draft depth grows. This problem becomes more severe in tree-based drafting, where distinct branches are forced to share the same marginal distribution for subsequent tokens. We propose TreeFlash, which addresses this by incorporating an MLP layer conditioned on the drafter's hidden state and the previous token to approximate an autoregressive distribution. TreeFlash retains the \(\mathcal{O}(1)\) decoding time complexity of one-shot drafters by employing a two-stage approximation mechanism. TreeFlash achieves state-of-the-art performance across a variety of tasks and models, improving over marginal tree drafting by \(12\%\) higher block efficiency and \(9\%\) higher speedup.

Quantization is a standard path to deploying large language models, and a quantized model is typically judged acceptable when its perplexity or downstream accuracy stays close to the full-precision original. Whether the model still computes in the same way, or whether the interpretable features identified in the full-precision model survive weight rounding, is rarely tested, even as safety audits and steering interventions increasingly rely on those features. We ask whether sparse autoencoder (SAE) features extracted from a dense full-precision model remain faithful once that model is quantized. Using a frozen SAE as a fixed measurement basis, we encode full-precision and round-to-nearest (RTN) quantized activations on identical tokens and quantify per-feature survival by Pearson correlation, sweeping bit-widths from INT8 to INT4 on Pythia-70M and Gemma-2-2B. We find that feature survival is graded: features degrade systematically rather than failing all at once, with 62.4 percent of active features surviving at INT6 on Pythia-70M and 51.3 percent surviving at INT6 on Gemma-2-2B, and with most non-survivors blurred rather than destroyed. Survival is predictable from full-precision statistics alone, with cross-validated AUCs of 0.92 to 0.97 and peak activation as the strongest marginal predictor. Critically, task metrics can miss this damage: on Gemma-2-2B, INT7 improves perplexity while degrading 18.7 percent of features. Finally, quantization and matched-perplexity magnitude pruning damage strongly overlapping feature sets, with Jaccard overlap of 0.79 to 0.86 and damage-score Spearman correlation of 0.98, suggesting a shared mode of compression-induced vulnerability. These results show that behavioral parity is insufficient evidence that interpretability findings transfer to quantized deployments, motivating feature-level audits of compression.

Large Language Models (LLMs) have become integral to modern applications, yet their deployment remains challenging. Beyond executing the models themselves, practical deployment must address cost efficiency, low latency, and optimal resource utilization. Conventional approaches typically assume that an entire model can be hosted on a single device, which does not hold in many real-world scenarios, particularly in Edge and Fog environments where device resources are constrained. In this paper, we introduce E2LLM, a framework designed to enable efficient LLM deployment in such resource limited settings. Rather than simply partitioning a single model across all available devices, E2LLM replicates the full model across multiple groups of devices (replicas) and applies model parallelism within each replica. Each replica is assigned a specialized role PREFILL or DECODER based on its efficiency in handling input and output tokens. This separation leverages the inherent differences between these two phases of LLM inference. To effectively organize devices, we utilize a Genetic Algorithm to form clusters that maximize system performance. Within each cluster, we apply Dynamic Programming to determine an optimal partitioning strategy that minimizes bottlenecks in model-parallel execution. Experimental results demonstrate that our approach adapts robustly to varying workloads, including scenarios with significant variation in input and output token lengths. Compared to the Splitwise baseline, E2LLM reduces average waiting time by over 50% under high-demand conditions

Disaggregated LLM inference forces the KV cache to traverse the datacenter network before decoding begins, so transfer time enters directly into the Time to First Token (TTFT) budget. Current schedulers route on compute load and prefix-cache locality alone, ignoring the topological distance and dynamic congestion between prefill and decode instances. We close this gap with a thin operator-to-scheduler interface, the network cost oracle, and we prove that ignoring the network term renders cache-aware-only scheduling arbitrarily suboptimal as context length grows. NetKV, the O(|D|) per-request greedy that consumes this oracle, has tier rankings that are provably robust to stale telemetry. On a 64-GPU four-tier fat-tree simulator driven by Mooncake traces, NetKV reduces mean TTFT by up to 21.2% over round-robin and 17.6% over a tuned cache+load-aware scheduler, lifts SLO attainment by up to 20.1 percentage points, and keeps the Time Between Tokens overhead below 0.5 ms in every condition tested, with no changes to the transport, inference engine, or hardware.

Reasoning models improve accuracy through extended chains of thought, but their long outputs create a memory and compute bottleneck. KV cache eviction methods reduce this cost by evicting unimportant key-value pairs from the cache, yet they often yield worse accuracy than selection-based sparse attention alternatives, which keep the full KV cache. We identify key factors crucial to KV cache eviction accuracy. First, a small fraction of value states have abnormally large magnitudes, and evicting them causes catastrophic failure where models enter repetitive reasoning loops. Second, introducing stochasticity during eviction improves accuracy by increasing cache diversity. Based on these findings, we propose Value-aware Stochastic KV Cache Eviction (VaSE), a training-free recipe that protects large-magnitude value states and promotes diverse eviction decisions. Across six reasoning tasks, Qwen3 models using VaSE with 4x KV cache compression yield higher average accuracies than SOTA selection method at the same sparsity, while outperforming the strongest eviction method by more than 4%. Overall, VaSE bridges the gap between efficiency and accuracy, supporting FlashAttention2 and enabling a static memory footprint for reasoning models.

Test-time scaling is a powerful approach to obtain better reasoning in large language models, but it becomes memory-bottlenecked during long-horizon decoding, as the KV-cache grows. KV-cache quantization can help improve this, but current methods are evaluated under prefill-like settings and errors behave differently under autoregressive decoding. We show that in the latter regime, quantization errors accumulate across timesteps, driven primarily by incorrect token scales. We introduce KVarN, a calibration-free KV-cache quantizer that applies a Hadamard rotation followed by a dual-scaling variance normalization across both axes of the K and V matrices. We find that this combination fixes outlying token-scale errors and substantially reduces error accumulation over existing baselines. KVarN establishes a new state-of-theart for KV-cache quantization on generative benchmarks, including MATH500, AIME24 and HumanEval, at 2-bit precision. A vLLM implementation of the KVarN method is available at https://github.com/huawei-csl/KVarN

We present AAD-1, an Asymmetric Adversarial Distillation framework for One-step autoregressive image-to-video generation. State-of-the-art methods adopt adversarial distillation but suffer from motion collapse and training instability, resulting in static videos. AAD-1 addresses these challenges through two key designs in architecture and training strategy. Our key architectural insight is to break the symmetry between generator and discriminator. While the generator remains causal to preserve autoregressive sampling capability, the discriminator attends bidirectionally over the full spatiotemporal context and produces a single holistic realism score for the entire video sequence. This asymmetric design enables the discriminator to effectively detect global temporal failures and long-range drift that cause motion collapse in autoregressive generation. To stabilize training, we introduce a phased strategy that first uses distribution matching to bootstrap a stable one-step generator, providing a warm-up phase that brings the student distribution closer to the teacher before adversarial distillation begins. Extensive experiments on VBench demonstrate that AAD-1 achieves state-of-the-art performance in one-step autoregressive video generation.

Large Language Model (LLM) inference relies on key-value (KV) caches to avoid redundant attention computation. While approximate KV cache retention techniques reduce memory usage by sacrificing model accuracy, lossless approaches instead evict KV cache blocks from GPU memory and reconstruct them on demand to preserve exact outputs. Existing lossless KV cache management systems primarily base eviction decisions on access frequency or positional heuristics, without considering how different KV cache blocks affect the execution efficiency of GPU attention kernels. In this paper, we propose AsymCache, a computation-latency-aware KV cache management system for LLM inference that explicitly aligns cache residency decisions with GPU attention kernel performance, including three key components: Multi-Segment Attention (MSA) for efficient non-contiguous KV context processing, a cache eviction policy that jointly optimizes hit rate and position-aware recomputation cost, and an adaptive chunking scheduler for high hardware utilization. Experiments show that AsymCache reduces TTFT by up to 1.90-2.03x and time-per-output-token (TPOT) by 1.62-1.71x over latest baselines, confirming the effectiveness of the method in common workloads and validating its design goal of balancing computational efficiency with cache hit rate. Moreover, the low-level design of AsymCache allows seamless integration into agent serving systems such as Continuum, where it further reduces average job latency by up to 18.1%.

Speculative decoding accelerates inference by having a lightweight drafter propose tokens verified in parallel by the target language model. Block diffusion drafters such as DFlash generate an entire draft block in one pass, yielding per-position marginals; DDTree uses these to build a candidate tree that maximizes expected acceptance length under a fixed node budget. We observe, however, that acceptance length is non-decreasing in budget: it always favors larger trees regardless of verification cost, offering no principled basis for budget selection. We introduce \textbf{CaDDTree} (Cost-aware Diffusion Draft Tree), a method that directly optimizes token throughput (expected tokens generated per unit time) by jointly selecting the tree structure and node budget. We model draft and verification latencies explicitly, show that the throughput objective decomposes into a per-round one-dimensional search over the budget, and prove that under a convex verification cost the throughput function is \emph{unimodal}, enabling an efficient greedy stopping rule. CaDDTree requires no offline budget search, adapting the budget each round from the current per-position distributions and verification cost. Experiments on Qwen3-4B and Qwen3-8B across eight benchmarks spanning reasoning, coding, and instruction-following tasks show that \caDDTree{} matches or surpasses DDTree with oracle budget selection on nearly all tasks.

Distribution Matching Distillation (DMD) compresses pretrained diffusion models into efficient few-step generators by aligning their noised distributions across all scales. In principle, such distribution-level supervision remains agnostic to specific noise-data pairings of the teacher; this provides the student the freedom to remap latent noise, a behavior consistently observed in low-dimensional settings. Surprisingly, we find that in high-dimensional settings, distilled students spontaneously reproduce the original noise-data pairings of the teacher, a phenomenon we term copying. We demonstrate that copying is neither a byproduct of adversarial objectives nor a result of teacher memorization. Instead, our evidence suggests that copying is an emergent property arising from the limited geometric freedom of the student model during high-dimensional distillation.