每日文章

Quantization is widely used to reduce the inference cost of large language models, but its effect on reasoning models is not fully captured by final-answer accuracy or per-token latency. We show that low-bit post-training quantization can introduce a hidden test-time compute cost: quantized reasoning models often generate longer chains of thought even when they still answer correctly. Across mathematical reasoning, code generation, scientific question answering, and agentic tool-use benchmarks, we find that INT4/INT3 quantization can preserve accuracy but increase reasoning-token usage, offsetting the expected per-token speedup. To measure this effect, we introduce the CoT Token Inflation Ratio, which compares reasoning length between quantized and full-precision models averaged across all evaluation benchmarks. We further show that token inflation is accompanied by behavioral changes in the reasoning trace, including more intermediate steps and greater semantic repetition. These changes translate into measurable end-to-end real-world serving penalties. Finally, we evaluate mitigation strategies and find that prompting and decoding-time sampling offer inconsistent accuracy-length trade-offs, while quantization-aware training shows more promise in reducing both accuracy degradation and token inflation. Our results suggest that reasoning-token usage should be reported alongside accuracy when evaluating quantized reasoning models.

Currently, streaming video understanding is still a daunting task for existing multimodal large language models (MLLMs). Its difficulties not only lie in handling the ever-increasing video frames, but also in the unpredictability of future video content and input instructions. In this paper, we study this task from the perspective of constructing a dynamic but fixed-budget memory bank, and propose a novel and training-free approach termed CausalMem. CausalMem is dedicated to constructing a dynamic visual memory update mechanism, thereby maximizing the amount of information in streaming video within a limited memory space, much like the human brain. In practice, CausalMem estimates the redundancy of visual tokens and updates the memory bank via an online semantic basis, which models the principal semantics of the observed video stream. To validate CausalMem, we apply it to two representative MLLMs, namely LLaVA-OneVision and Qwen2.5-VL respectively, and conduct extensive experiments on both streaming and offline video understanding benchmarks. The experimental results not only show the great advantages than existing methods under both streaming and offline settings, e.g., \(+3.2%\) and \(+3.0%\) average accuracy gains respectively, but also witness the superior semantic preservation for streaming videos, e.g., using 12\(k\) token budgets to memorize hour-long streaming videos, which achieves more than 20\(\times\) visual token compression ratio and only occupies about 82 MB storage. Our code is given in CausalMem.

On Apple Silicon the fp32 GEMMs dominating LLM prefill are dispatched by Accelerate to a matrix coprocessor (AMX) on the M1-M3. We ask where a hand-written kernel's throughput over Accelerate comes from on the M1 AMX, and reach a structural conclusion: not a faster inner loop. By microbenchmark the inner loop is load-issue bound -- once any operand load interleaves with the FMA32 stream, single-thread throughput falls to a 610-to-680 GFLOPS band, under half the load-free rate. The gain comes from two deployment-level levers Accelerate underuses: fine multi-thread panels filling the M1's second on-chip AMX block (winning the K >= N shapes), and pre-packing the constant weight at load (winning the N > K shapes). A bit-exact direct-AMX kernel using both is the fastest bit-exact fp32 GEMM path we find on the M1: it exceeds all three Accelerate fp32 paths (cblas_sgemm, BNNSMatMul, and the BNNS Graph compiler) at all twelve LLM prefill GEMMs at S = 128 (GPT-2 to Llama-7B scale), leading the fastest, BNNS Graph, by 1.17 -- and by 1.09 at the three shapes where it too holds fp32 -- with geometric means of 1.58 over BNNSMatMul and about 2.0x over cblas_sgemm. Every output is bit-identical to Accelerate, whereas BNNS Graph is bit-exact at only three of twelve shapes, the rest at reduced precision (error up to 1.4e-3). Dropped into llama.cpp for its cblas_sgemm prefill matmuls, it raises measured full-forward throughput from 291 to 420 tokens/s (1.44x, bit-identical) at 128-token prefill -- end-to-end, not a GEMM-only ratio. The contribution is this shape-resolved M1-AMX characterization (microbenchmark, two-block aggregate, per-core occupancy probe), leaving fine-panel scheduling and pre-packing as the only two levers above an inner loop at the hardware limit; mis-tuning the single column-panel width costs nearly 2x.

Lifelong continual learning remains an obstacle on the path to human-like intelligence. Modern transformers show sparks of intelligence with in-context learning. The quadratic nature of attention, however, prohibits transformers from performing this process on arbitrarily long sequences. In this work, we argue that extending in-context learning to lifelong settings is a practical solution for continual learning in AI agents. In particular, we argue that parametric forms of attention are needed to understand a lifetime of context with transformers on a fixed hardware budget. These attention mechanisms learn the relationship between keys and their associated values at test-time with parametric regression. Our generalization of parametric approaches (linear attention, state-space models, fast weight programmers, and test-time training layers) contrasts with nonparametric counterparts like softmax attention. They replace the ever-growing key-value cache with an online-trainable neural network, maintaining a constant memory footprint. We highlight how parametric attention currently fall short of lifelong learning due to limited memory capacity or costly online updates. To address these issues, we pose a set of open questions with novel insights to guide the field toward long-horizon agents.

General-purpose robots operating in partially observable environments, such as homes, require memory to support autonomy. They must recall diverse information from the past, such as where objects were placed, which tasks a human partner has completed, and when an appliance was turned on. Achieving this versatility requires a general memory retrieval mechanism. Transformer architectures that use attention over long contexts for memory retrieval provide a promising approach, as they learn retrieval from data rather than relying on task-specific or hand-designed rules. However, directly incorporating them into imitation learning from offline data introduces two key challenges: (1) the policy may learn spurious correlations between past information and predicted actions, and (2) errors accumulate in memory due to prediction inaccuracies and their compounding interactions with the environment, causing model drift and cascading failures. To address both challenges, we introduce HALO, a visuomotor policy with an attention-based memory retrieval mechanism for long-horizon control. First, to suppress spurious correlations, HALO distills vision-language model (VLM) priors into the policy. It generates memory-dependent question--answer pairs from demonstration trajectories and trains jointly with a video question--answering objective, steering retrieval toward task-relevant information. Second, to reduce the impact of accumulated errors in memory during closed-loop control, HALO uses sparse attention that restricts retrieval to only the most relevant parts of the history. Together, these components enable more reliable long-horizon control by guiding the policy to retrieve task-relevant information from up to eight minutes of past experience. Project website: https://robin-lab.cs.utexas.edu/HALO

Speculative decoding (SD) accelerates LLM inference by \(1.5\)-\(3\) times when the draft and target models are co-located. This has motivated a distributed variant (DSD) that places the draft model on an edge device while the target stays in the cloud. We show with closed-form inequalities that DSD's per-request latency benefit is limited under WAN edge-cloud communication. If the server can host both models, co-located SD has lower latency and communication than synchronous DSD, with the same per-output FLOPs and model-weight memory. Pipelining can make DSD competitive with co-located SD only in low-RTT regimes where the round trip is shorter than the edge drafting time window; at WAN RTTs, the cloud round trip remains too large for pipelined DSD to beat co-located SD. Against cloud autoregressive decoding, DSD can reduce latency only inside a bounded window given the target-model speed, acceptance rate, and RTT. DSD is also infeasible against closed-source APIs without a verifier-only interface. The main case for DSD appears in multi-tenant capacity. Under cross-client overlap, offloading draft compute lets a saturated cloud server sustain \((1 + γ\,t_d/t_v)\) times more concurrent clients at the same per-client rate, where \(γ\) is the speculation length and \(t_d, t_v\) are the per-step draft and verification times. DSD should therefore be evaluated primarily by multi-tenant capacity and server throughput, not only by single-request latency.

While speculative decoding improves inference throughput for multi-batch long-context Large Language Models (LLMs), its efficiency is often limited by a verification bottleneck where Key-Value (KV) cache loading dominates latency. Existing compression methods fail in this regime: static eviction incurs accuracy loss due to saliency shift, while dynamic selection introduces prohibitive computational overhead during the verification path. We propose Dustin, a sparse verification framework designed for long-context speculative decoding. Dustin integrates lookahead signals from the draft model with historical attention from the target model to identify critical tokens with high fidelity across multi-step verification windows. To reduce recomputation latency, this approach further employs a sparse estimation scheme that restricts importance scoring to a minimal subset of attention heads. Evaluations on PG-19 and LongBench with Qwen2.5-72B demonstrate that Dustin achieves a 27.85x speedup in self-attention and a 9.17x end-to-end decoding speedup at a 32k sequence length, all with negligible accuracy degradation.

Speculative decoding accelerates inference by letting a draft model propose tokens for a target model to verify, raising a concrete safety question: at temperature zero, can draft-side behavior leak into safety-scored outputs? We answer with Typical-Acceptance Invariance Screen (TAIS), a behavioral-equivalence screen that pairs target-only and speculative outputs on the same safety battery and requires byte-identity evidence, TOST equivalence at +/-3pp, and per-task Cohen's h below a calibrated null cutoff of |h| < 0.1. Applied to a 16,783-sample confirmatory core plus 44,066 matched expansion samples (fp16/bf16 execution, canonical and DPO-adversarial drafts, GPTQ-4bit drafts, two seeds, and four safety benchmarks), the tested temperature-zero vLLM stacks show no detectable safety divergence under TAIS. The largest absolute Cohen's h on matched target-only versus speculative refusal is 0.024, roughly an order of magnitude below the conventional trivial-effect floor; 25 of 27 per-task TOST contrasts pass at the +/-3pp margin (the two non-pass contrasts are capability-domain Wald-CI edge cases at identical ceiling rates, not genuine non-equivalence); the DPO-adversarial draft produces byte-identical output to the canonical draft across 4,006 samples; and bf16 changes 36%-53% of output bytes without moving any per-task safety rate outside equivalence. A separate 4,006-sample 70B production-scale probe, which lacks a matched 70B target-only arm and is therefore not counted as a TAIS pass, produces AdvBench refusal 0.839 over 700 AdvBench completions with 95% Wilson CI [0.809, 0.864]. We make no claim about sampling temperatures, untested frameworks, untested model families, or tree-speculation variants such as EAGLE and Medusa.

Traffic accident reconstruction is a forensic inverse problem that requires recovering physically consistent motion from sparse and heterogeneous evidence. Existing learning-based approaches predominantly optimize for semantic plausibility or visual realism, rather than quantitative agreement with measurable geometry and dynamics. Here, we present TRACER, a training-free framework that formulates reconstruction as a closed-loop structured inference process. Instead of directly generating dense trajectories, our framework constructs and iteratively refines event-anchored motion hypotheses under geometric, kinematic, and interaction constraints, guided by structured case memory and consistency-driven diagnosis. This design enables incremental, interpretable corrections when evidence is insufficient, making the accident reconstruction process more aligned with the workflow of human experts. Experiments on real-world accident data show that TRACER achieves improved geometric fidelity, velocity consistency, and collision accuracy over both data-driven and physics-based baselines.

Sparse Mixture-of-Experts (MoE) models route each token to a few of many experts, inviting the hypothesis that experts form functional modules tied to capabilities or languages. We test this causally on Command A+, a frontier open-weights MoE (218B total / 25B active; 128 experts, 8 active, +1 shared). We build a routing-mass atlas, pre-register six family-to-axis hypotheses before any intervention, and ablate each family at inference time against a size-matched random-expert null, measuring whether it selectively breaks its own axis (worst off-target effect at most one third of on-target). Crucially, we test the same families under four metrics and a held-out, independent-corpus run with bootstrap confidence intervals. Our finding is cautionary: robust functional modularity is rare and measurement-dependent. Of six pre-registered families, only one, the Arabic-language family, is a clean selective module that survives an independent corpus and a conservative statistical bar (1/6; a more permissive pre-registered point rule admits 3/6, but that count is threshold-sensitive). Every other family has a real causal effect yet fails selectivity, and its apparent modularity flips with the measurement: with the corpus, the metric, and the statistical bar. A positive control on Qwen3-30B-A3B recovers its published disjoint structure, confirming the method detects modularity when present. The verdict reproduces on the un-quantized BF16 model, ruling out a 4-bit quantization artifact. We conclude that ablation-based modularity verdicts are not safe unless the corpus, metric, and statistical bar are controlled. We release the atlas and ablation data.

The accelerated digitalization of renewable energy smart grids through IoT sensors, AMI, and SCADA systems has significantly expanded the attack surface for sophisticated cyberattacks, FDI attacks that stealthily distort state estimation and DoS/DDoS attacks that flood communication channels. Current IDS, however, exhibit three inherent limitations: inadequate modeling of the temporal progression of multi-step attacks, degraded scalability under extremely skewed class distributions of standard benchmark datasets, and restricted generalization across heterogeneous network environments. In this study, we present a Hybrid CNN-LSTM IDS that jointly exploits CNN-based spatial feature extraction and LSTM-based temporal sequence modeling, enabling the detection of instantaneous volumetric anomalies and gradually evolving low and slow-attack campaigns in real time. The model was trained using a seven-step preprocessing workflow comprising missing-value imputation, min-max normalization, one-hot encoding, SMOTE class balancing, mutual-information feature selection, causal temporal sequence construction (T=10), and stratified partitioning. LSTM (96.1%), Random Forest (93.5%), SVM (91.2%) and KNN (89.7%); in NSL-KDD, it reaches 98.2% precision versus 96.4% (LSTM), 95.2% (CNN), 92.7% (Random Forest) and 90.8% (SVM), with margins of 2-9 percentage points in all measures. An ablation analysis identified SMOTE balancing as the most influential design choice (-3.7~pp F1 without it). The model achieves a real-time inference throughput of 27,800 flows/s on GPU and 0.082 ms/sample CPU latency in FP32,, with INT8 quantization providing an additional 3.1 x speedup at 0.3% accuracy loss, confirming deployment feasibility on resource-constrained IEDs with <128MB memory and establishing a deployable deep-learning framework for securing next-generation renewable energy smart grid infrastructure.

Most vision-language-action (VLA) models are reactive: they predict the next action from the current instruction and observation, implicitly assuming that the current observation fully specifies the action-relevant state. In embodied control, however, embodiment-specific factors such as camera-to-robot geometry, robot calibration, or systematic actuation bias are often hard to identify from a single observation. As a result, reactive policies cannot reliably disambiguate these factors in general, overfitting to training environments and generalizing poorly at deployment. We propose Reflective VLA, which conditions each decision on a context of observation-action-consequence triplets. Each triplet records not only what the robot observed and executed, but also how the scene changed afterward, exposing the deployment-specific mapping from actions to observed effects. Architecturally, Reflective VLA routes all observation modalities through the VLM under shared attention, so the action expert reasons directly over past triplets and the current observation. A block-causal mask enables parallel multi-frame training without leakage and supports KV-cached real-time inference. On standard LIBERO and SimplerEnv-Bridge, Reflective VLA preserves strong in-distribution performance. Under distribution shift on LIBERO-Plus and the harder LIBERO-Plus-Hard, it improves average success rate by 5.4 and 4.2 percentage points over a matched reactive baseline. Ablations with a matched history-only baseline further show that action consequences -- rather than additional context length alone -- are the key to cross-environment generalization. Project page: https://lianqing11.github.io/reflective-vla-page/

Serving diffusion models for image-to-video generation is computationally expensive, posing significant challenges for large-scale deployment. Real I2V workloads often contain similar requests, such as repeated effect templates, related subjects, and recurring shot layouts. Existing cross-request acceleration methods mainly exploit this redundancy through feature reuse. We observe that similar I2V requests also share highly consistent sparse attention patterns, enabling historical sparse masks to serve as request-conditioned priors with almost no online mask-prediction overhead. We propose a cross-request reuse framework centered on sparsity reuse, with feature reuse as an optional extension safeguarded by a lightweight guidance enhancement. Our sparsity reuse is implemented as shared sparse mask reuse, which reuses high-quality sparse masks from similar historical requests to avoid per-request online mask prediction. Optional feature reuse applies downsampled computation to highly redundant spatiotemporal regions, mitigating boundary artifacts while preserving efficiency gains. Guidance enhancement reinforces image/text conditioning after reuse, mitigating semantic drift and condition-adherence issues. Experiments show that default sparsity reuse configuration preserves generation quality with a 2.16\(\times\) speedup.

Target-specific peptide design requires sequence and structure co-design under full atom geometric constraints. Latent generative frameworks offer an effective route for this problem by compressing fine grained atomic structures into block level latent representations and performing conditional generation in a compact latent space. However, the scalability of such systems depends heavily on the geometric backbone used throughout their encoding, decoding, and denoising components. We introduce MEET (Memory Efficient Equivariant Transformer), an E(3) equivariant backbone for scalable atomistic peptide modeling. MEET maintains coupled invariant scalar and equivariant vector feature streams, while reformulating geometric computation around memory efficient attention. It initializes vector features through global coordinate aggregation, incorporates pairwise distances through augmented query and key dot products, and injects covalent bond information through sparse bond adaptation. Integrated into a VAE and latent diffusion pipeline for full atom peptide generation, \model{} achieves linear memory scaling with atom count and improves generation quality over existing peptide design methods. Experiments on large scale AFDB derived datasets further show that the proposed backbone supports systematic model and data scaling, leading to better binding affinity, physical validity, and sample diversity.

Modern large language models based on softmax scaled-dot-product attention are constrained by their training sequence length: as the key-value sequence grows, softmax probability mass can dilute across a wider distribution, inducing activation shift and long-context performance collapse. Moreover, long-context language modeling faces a structural tension: a sliding-window attention core maintains a bounded local representation and low perplexity but is blind to long-range dependencies, while full-context attention preserves global recall but suffers from out-of-distribution perplexity explosion. To resolve these limitations, we introduce ATMA, a hybrid convolutional-attention architecture that integrates a novel three-channel attention mechanism. ATMA factorizes the attention mixing step into: (1) a count-blind, unit-vector direction channel, (2) a bounded magnitude channel driven by the participation ratio of effective matches over an extreme-value-corrected null sink, and (3) a long-term recurrent compression memory optimized via a gated-delta fast-weights rule. Neither the Polar Attention core nor the recurrent memory is sufficient alone; their combination enables monotonic perplexity reduction and high-fidelity long-range retrieval simultaneously. We evaluate ATMA using a 100-run factorial ablation sweep, demonstrating that the combined Polar + memory model maintains induction needle-in-a-haystack retrieval accuracy above 90% out to 64K tokens (32 times the training length of 2K) while its document perplexity improves monotonically, outperforming softmax-based memory baselines which collapse at extreme context lengths. Code: https://github.com/kreasof-ai/atma

Pruning Large Language Models (LLMs) reduces memory and inference costs by removing parts of the network, producing smaller models that retain most of their accuracy. As attention layers are the most resource-intensive parts of LLMs, pruning them is a promising compression strategy. Prior work shows that up to 33% of attention layers can be pruned with minimal accuracy loss. Nevertheless, the impact of attention pruning on model interpretability, specifically faithfulness and confidence calibration, remains unstudied. To address this gap, we study how pruning attention layers affects explanation faithfulness and confidence calibration across five LLMs and eight datasets. While the pruned models often maintain high accuracy, we find that their faithfulness and calibration often degrade. Notably, faithfulness and calibration can fluctuate significantly, even when accuracy remains stable, highlighting a misalignment between model confidence, interpretability, and accuracy. Our findings suggest that layer pruning can affect LLMs' interpretability and reliability in ways not captured by accuracy and efficiency measures alone. We recommend including explainability and calibration metrics when evaluating pruned models.

Light field (LF) image denoising is challenging due to the high-dimensional structure of LF data. While noise is independent across sub-aperture images, scene content exhibits strong cross-view correlations. We introduce VSANet, a view-aware sparse attention network for LF denoising. Specifically, we propose a view-aware sparse attention (VSA) block that represents the 4D LF feature map as a unified spatial-angular token space and performs cross-view aggregation via locality-sensitive hashing-based sparse attention. This enables global feature interactions with linear complexity, effectively exploiting LF correlations across views and spatial locations. In addition, we design a feature refinement (FR) block to emphasize informative features in spatial, angular, and epipolar subspaces. The VSA and FR blocks are integrated within a sequential attention refinement module, forming the core of VSANet. Experiments demonstrate VSANet outperforms stateof-the-art LF denoising methods.

Quadratic unconstrained binary optimization (QUBO)-based quantum computed tomography (CT) casts reconstruction as a binary quadratic problem for quantum annealing and hybrid quantum--classical solvers. For grayscale CT, however, image encoding is constrained by the binary-variable budget: fixed global bit-plane encodings increase QUBO size and coupling complexity as gray-level precision improves, whereas low-bit encodings introduce quantization error. We propose a QUBO-based grayscale CT reconstruction framework that combines dynamic interval encoding with prior-balanced optimization. Each refinement round encodes active pixels only within local gray-level intervals around the current estimate, and a boundary-hit-guided update rule adaptively switches between search expansion and local refinement. To improve optimization stability, the method balances projection-domain data consistency and an edge-preserving quadratic prior before forming the final QUBO. Sparse-view and limited-angle fan-beam CT experiments show that the proposed method recovers structures and gray-level distributions more faithfully than the evaluated analytic, iterative, variational, and representation-based baselines. Expressivity analysis and ablation studies further indicate that the improvement mainly arises from effective gray-level representation through dynamic local encoding and more stable data-fidelity--prior coupling. Experiments on the D-Wave hybrid binary quadratic model (BQM) solver further demonstrate that the formulation is executable on a hardware-backed hybrid quantum--classical backend.

Memory stranding wastes 25-35% of installed DRAM in production cloud clusters. Memory pooling over CXL and RDMA offers a remedy, but neither technology alone suffices: CXL provides low-latency, load/store-transparent access limited to a pod, while RDMA provides cluster-wide reach at higher latency with software overhead. A hierarchical architecture combining both tiers is the practical path forward, yet remains unexplored for MicroVM-based serverless computing, where snapshot restore latency is the dominant cold-start bottleneck. We present Aquifer, the first system to serve MicroVM snapshots from a hierarchical CXL+RDMA memory pool. A characterization of snapshot images reveals that the vast majority of pages are either zero or cold, enabling a hotness-based snapshot format that eliminates zero pages and places only the hot working set in the CXL pool while storing cold pages in the RDMA pool. Sharing these snapshots across hosts on CXL 2.0 multi-headed devices, which lack hardware cache coherence, requires Aquifer's ownership-based coherence protocol to ensure correctness. Finally, Aquifer uses a copy-based page serving mechanism pre-installs hot pages from CXL memory before MicroVM resume and demand-pages cold pages asynchronously from RDMA. On emulated CXL+RDMA hardware, Aquifer achieves a 2.2x geometric-mean speedup in end-to-end invocation time over Firecracker and 1.1x over the next best alternative.

Adapting a streaming speech recognition model to a new language requires choosing between two plausible warm starts: a multilingual (ML) encoder or an English-only (EN) encoder. The common intuition is that the multilingual encoder should help most at low data, but it is unclear how long that advantage persists, whether tight streaming latency amplifies it, and whether it survives deployment quantization. We answer these questions with a controlled sweep of a 0.6 B-parameter cache-aware FastConformer transducer across eight European languages, up to five target-language data scales (100 h to 2500 h), three streaming tiers plus offline decoding, and up to four public test sets. The main result is that multilingual initialization is a data-limited advantage, not a latency-limited one. On FLEURS at 160 ms, the mean EN-ML word error rate (WER) gap falls from +4.21 percentage points (pp) at 100 h to +0.20 pp at 2500 h; a power-law fit summarizes this decay, with each doubling of target-language data roughly halving the remaining advantage. Across the three streaming tiers, the across-language mean EN-ML gap is approximately stable at each scale from 100 to 1000 h, and is near zero by 2500 h. Finally, 4-bit weight-only encoder quantization at the matched 560 ms streaming tier reduces the encoder footprint by about 3x, with an average FLEURS WER increase of about 0.5 pp. The resulting guideline is simple: use multilingual initialization in low-data regimes, treat the choice as effectively irrelevant at large data, and make latency and quantization decisions independently.