跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+479新增论文
+361其中 AI 相关
+63已有论文修订
+93新增 AI 导读

arXiv 最近完成扫描:2026-11-16 01:34 · 全站最近入库:2026-11-16 13:44各信源进度 →

统计口径与数据新鲜度

统计 2026-11-15 18:06 — 2026-11-16 18:06(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

最近完成检查:2026-11-16 01:34最近发现新增版本:2026-11-16 13:44库内论文最新更新时间:2026-11-14 02:59

每分钟刷新已保存的数据。空检查不代表采集异常;新收录可能包含历史论文。AI 导读按新增版本计数。

23,827 已收录论文19,941 AI 相关6,921 有修订112 HF 社区发现

浏览论文库

时间范围跟随排序:版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常,查看状态 →

找到 23,827 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-16 01:33(北京时间)
新论文 · v1arXiv:2610.39882

LAMP: LLM-Driven Analog Placement through Adaptive Tool Composition

Zecheng Xu, Guanhua Chen, Yumao Wu, Keren Zhu

原文摘要

Analog placement requires circuit-specific trade-offs among area, matching, routability, and electrical performance. Balancing these trade-offs through predefined objectives and fixed placement flows remains difficult. In this paper, we propose an analog placement framework in which a large language model (LLM) acts as a layout engineer through composable placement components. These components support constraint management, placement and legalization, and lightweight evaluation of routability and electrical performance. Using multimodal checkpoints of the initial and evolving design, the LLM selects and combines components to construct and revise its own placement flow. It also adapts the placement hierarchy by reopening modules or partitioning and regrouping devices, allowing local layouts to respond to global design needs. An experience skill provides reusable design knowledge to guide these decisions. Evaluation on two SMIC 180 nm circuits shows that the framework reduces placement area and improves most measured post-layout performance metrics.

cs.AR

新论文 · v1arXiv:2610.39880

What Survives Label Downsampling? Exact Region Survival for Coarse Segmentation Targets

Zhiyuan Zhang, Yukun Ou, San Tek Choi

原文摘要

Segmentation networks that predict at an output stride are trained against label maps downsampled by strided sampling, nearest-neighbor resampling or majority pooling. These methods label each coarse grid cell on its own, so a region that wins no cell is absent from the target and receives no positive label. We present MatchDown, which builds the coarse target from a minimum-cost matching between regions and the grid cells they occupy. It keeps as many regions as any one-region-per-cell target can, certifies how many losses are unavoidable, and with priority tiers (MatchDown-P) keeps every evaluated region (a region that a metric scores) of at least one cell's area. On COCO, Cityscapes and BBBC038 at factors 2 to 16, MatchDown-P loses exactly the minimum number of regions, whereas the downsampling used by StarDist, nnU-Net and connectomics pyramids loses four to five times as many on COCO at factor 8. Keeping a region in the target and reconstructing its shape are different problems: we show that no region lost by majority pooling can be reconstructed above IoU 0.5 by upsampling any coarse target, yet on BBBC038 a star-convex head trained on MatchDown targets segments 35% of the nuclei that majority pooling loses above IoU 0.5, against 15% when trained on majority targets and 24% on StarDist's strided target. On two nucleus datasets, with thresholds cross-validated for count error, this head has instance accuracy at IoU 0.5 (Acc50) 2.6 to 4.2 points above the strided target and at most 2.8 points below cell-center sampling, and a lower count error than the strided and cell-center targets on BBBC039. On BBBC039, an affinity head, which outputs one label per cell, lowers count error relative to majority targets by 9.3 nuclei per image at 7.2 points lower Acc50. MatchDown runs once, before training, in place of the label-downsampling step.

cs.CV

新论文 · v1arXiv:2610.39879

Compressed Orientation Reachability Maps for Sub-Second Robotic Needle Entry Point Planning

Prabhu Nithin Gollapudi, Andreas Maier, Franziska Mathis-Ullrich, Water Zhao, Jia Wang, Siming Bayer

原文摘要

In the context of multi-stage trajectory planning for robotic percutaneous liver interventions, identifying a kinematically feasible entry point remains a critical computational bottleneck. Existing methods rely on exhaustive inverse kinematics or randomized searches, incurring delays of 13 to 27 seconds per target and limiting real-time intraoperative adaptation. This paper introduces an Orientation-Based Reachability Map (ORM) that precomputes the kinematic feasibility of a 6 Degrees of Freedom (DoF) serial manipulator across 188,873 workspace positions and encodes orientation feasibility into a compact, indexed 83 MB map. A Just-In-Time (JIT) compiled runtime pipeline replaces live optimization with near-instantaneous array lookups. Experimental evaluation demonstrates a reduction in planning latency from 13.5-26.9 s to 0.289-0.349 s per target while maintaining high geometric fidelity, achieving a mean convex hull volume overlap ratio of 0.933 and a mean alpha-shape volume ratio of 0.976 relative to the baseline reachable point cloud. The proposed ORM framework enables sub-second entry point planning and provides an efficient computational framework for robotic needle insertion in percutaneous liver ablation.

cs.RO · cs.CG · eess.SY

新论文 · v1arXiv:2610.39876

Turing LLM Post-Training: Calibrated Distillation and Efficient Dynamic-MoE Execution

Yue He, Jiahui Hu, Shaman Tang, Hua Zhou, Qiman Wu, Menglin Li, Yizhao Wang, Yuheng Zhang 等 22 位作者

原文摘要

We study post-training for dynamically routed mixture-of-experts (MoE) models, focusing on integrating independently optimized specialists while keeping consolidation efficient. Our approach combines calibrated multi-teacher on-policy distillation (C-MOPD), which matches each selected teacher on a local Top-$K$ support with teacher-wise loss normalization and clipping, with an execution path that uses post-capacity bounded-width dispatch, final-route replay, and asynchronous rollout--training overlap. On Turing-20B-A2B, the selected model achieves a four-domain normalized integration score of 0.9988 and retains at least 90.16\% of the measured specialist gain in every domain. At matched $K=32$, the full calibration objective scores 0.9962, compared with 0.9724 without calibration and 0.9834 without clipping. At a matched 8K response cap, the model exceeds both shared SFT and Qwen3-8B on all four reasoning-enabled mathematics and code benchmarks (unweighted mean: 85.92, versus 80.33 for Qwen3-8B and 74.21 for SFT), and improves over SFT on ten of twelve broader evaluation metrics. In a PPU-based profile, the combined execution configuration reduces consolidation step time from 398.7 to 139.3 seconds, an observed 2.86$\times$ speedup over serial execution. These results demonstrate strong specialist integration alongside more efficient consolidation, while reflecting the complete training and execution framework rather than any single component.

cs.LG · cs.AI

版本更新 · v3arXiv:2409.16052

Denoising Graph Super-Resolution towards Improved Collider Event Reconstruction

Nilotpal Kakati, Etienne Dreyer, Eilam Gross

原文摘要

In preparation for Higgs factories and energy-frontier facilities, future colliders are moving toward high-granularity calorimeters to improve reconstruction quality. However, the cost and construction complexity of such detectors is substantial, making software-based approaches like superresolution an attractive complementary tool. This study explores integrating superresolution techniques into an LHC-like reconstruction pipeline to effectively enhance calorimeter granularity and suppress noise. We find that this software preprocessing step significantly improves reconstruction quality at fixed detector granularity, narrowing the performance gap to a finer calorimeter without closing it. To demonstrate its impact, we propose a novel transformer-based particle-flow model that offers improved particle reconstruction quality and interpretability. Our results serve as a proof of concept, demonstrating the potential of superresolution to enhance reconstruction in collider environments.

hep-ex · cs.LG

版本更新 · v5arXiv:2605.18267

SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation

Longtao Jiang, Jianmin Bao, Zhendong Wang, Xin Tao, Zhihui Li, Xiaojun Chang

原文摘要

Normalizing flows (NFs) provide exact likelihoods and deterministic invertible sampling, but have historically lagged behind diffusion models for large-scale image generation. We identify a key obstacle: NFs are required to learn a single invertible transport over the full ambient space, making them highly sensitive to high-dimensional representations. This leads to a semantic-capacity mismatch in modern visual representation spaces, where semantic information is compact but encoded in overcomplete features. We propose SRC-Flow, which introduces a Semantic Representation Compressor (SRC) to compact high-dimensional RAE features into a low-dimensional semantic space before flow modeling and preserve reconstruction through the frozen RAE decoder. This compact space reduces the modeling burden of NFs and enables effective likelihood-based generation in semantic representation space. We further adopt constant noise regularization tailored to the fixed unconditional bijection learned by flows. On ImageNet 256x256 and 512x512, SRC-Flow achieves state-of-the-art generation quality among normalizing flow methods, with gFID scores of 1.65 and 2.07 under classifier-free guidance, while retaining exact likelihood computation in the compact semantic representation space and deterministic invertible sampling at the flow level. Codes and models will be released in the future.

cs.CV

版本更新 · v2arXiv:2610.32512

Fast and Useful: Adapting Distilled Generators for Visual Learning

Patryk Bartkowiak, Dominik Michels, Sören Pirk, Wojciech Pałubicki

原文摘要

Distilled text-to-image models generate images quickly, but classifiers trained on those images can perform substantially worse than classifiers trained on images from their slower base models. We adapt released distilled generators to produce better training data while retaining fast generation. We fit a low-rank update to the student's first prediction so that it reaches an intermediate state produced by its unguided base, then disable the update and let the unchanged student finish generation. The base supplies training targets and is absent at inference. Guided targets and shorter jumps give classifiers lower accuracy in our ablations, showing why the state used for supervision matters. Across three generator families, the adapted images train better classifiers than images from the released students. With class-name templates at full ImageNet-1k scale, ResNet-50 top-1 rises from 8.4% to 34.2% for FLUX.2-klein and from 12.7% to 32.1% for SD-Turbo, using two and three denoiser evaluations per image, respectively. Fixed adapters also improve results on additional classes, fine-grained datasets, and other learner architectures and training recipes. Template-based adaptation remains useful against richer-prompt alternatives, while the benefit of combining adaptation with LLM generation prompts depends on the generator. Adapting a single generation step makes fast generators substantially more useful for training visual models.

cs.CV

新论文 · v1arXiv:2610.39873

Syndrome Trellis Decryption in Post-Quantum Cryptography

Meir Ariel

原文摘要

We present a novel post-quantum cryptographic framework based on masked high-memory convolutional codes in the Niederreiter paradigm. The public key is derived from a high-memory polynomial parity-check matrix augmented by dense random masking and additional invertible linear transformations. The resulting matrix exhibits strong random-like properties, effectively concealing its algebraic structure and resisting known structural and algebraic attacks. Legitimate recipients retain polynomial-time decoding capability, whereas adversaries appear to be limited to generic information-set decoding with exponential complexity. Under standard cryptanalytic estimates, the proposed construction achieves security margins exceeding those of classical McEliece- and Niederreiter-based systems by factors greater than 2^200. Beyond its enhanced security profile, the construction offers considerable design flexibility, supporting arbitrary plaintext lengths, linear-time decryption, and a uniform per-bit computational cost, enabling efficient scalability to very long messages. Practical implementation is facilitated by a parallel array of syndrome-trellis decoders, dynamically instantiated according to candidate syndrome values, enabling efficient hardware and software implementations with high throughput. Overall, the proposed scheme is a promising candidate for robust, scalable, quantum-resistant public-key cryptography.

cs.CR

版本更新 · v6arXiv:2502.08005

DiffRatio: Training One-Step Diffusion Models Without Teacher Supervision

Wenlin Chen, Mingtian Zhang, Jiajun He, Zijing Ou, José Miguel Hernández-Lobato, Bernhard Schölkopf, David Barber

原文摘要

Score-based distillation methods train one-step diffusion models in two stages: they first train a teacher score model, then distill it into a one-step student model. However, this distillation process can introduce bias from two sources: errors in the teacher score model and student score estimate. We propose DiffRatio, a new approach for training one-step diffusion models without teacher supervision based on a density-ratio formulation that connects score-based distribution matching and GAN-style adversarial training. Instead of using a teacher score model to provide training targets, DiffRatio directly learns a log density ratio between the student and data distributions across diffusion time steps. As a pre-trained score model is only used to initialize the one-step generator and does not supervise training, this design simplifies the training pipeline, mitigates gradient estimation bias, and reduces the size of auxiliary networks. The learned density ratio can be used as a verifier, enabling a principled inference-time parallel scaling method that further improves sample quality without external rewards or extra sequential computation. DiffRatio achieves strong one-step generation results on CIFAR-10 and ImageNet (64x64 and 512x512), outperforming most teacher-supervised distillation methods.

cs.LG · cs.CV

新论文 · v1arXiv:2610.39872

STORY2DIFF: A Dataset Linking Real-World User Stories to Code Changes

Benedetta Donato, Daniele Franchini, Alessandro Greco, Oliviero Riganelli, Daniela Micucci, Leonardo Mariani

原文摘要

Research on requirements-to-code tasks needs examples in which a genuine requirement is paired with the change that implemented it. Existing user-story collections, however, largely contain isolated or synthetic text. To address this limitation, we present STORY2DIFF, a dataset of 7,785 English user stories linked to merged pull requests from January 2017 to July 2026. The dataset is mined from a set of 982 public Java and Python repositories. Each record in the dataset combines the original issue, extracted Connextra elements, the associated pull request and commit metadata, a file-level diff capturing the code changes introduced by the associated pull request, and the affected files before and after the change. The tool we built to mine STORY2DIFF can be used to repeat the automatic extraction of the records, using different parameters (e.g., a different timeframe). The dataset can support a variety of tasks, including requirement discovery, traceability, change- impact estimation, and requirements-aware code generation. Dataset: https://github.com/riganelli/user-story-mining-pipeline/ releases/tag/V4.0.1. Video: https://youtu.be/qOhXUeyeiTQ.

cs.SE

版本更新 · v2arXiv:2606.10852

JANUS: Evaluating Goal-Conditioned Distortion in Factually Grounded LLM Communication

Polydoros Giannouris, Mohsinul Kabir, Sophia Ananiadou

原文摘要

LLM deception is often evaluated through fabricated claims, explicit lies, or strategic concealment. Yet misleading communication can also arise from selectively presenting true facts, for example, by omitting adverse evidence, softening unfavorable details, emphasizing favorable information, or replacing precise qualifications with vague language. Existing benchmarks largely overlook this subtler failure mode. We introduce JANUS, a benchmark for measuring goal-conditioned pragmatic distortion in fact-grounded LLM outputs. Each of its 160 scenarios, spanning 8 domains, provides a fixed pool of favorable and adverse facts and contrasts a neutral decision support with institutional conditioning, such as increasing adoption, or support despite potential harm to affected individuals or groups. By constraining outputs to the same fact pool, JANUS isolates misleading net impressions from hallucination and fabrication. Experiments across 12 LLMs reveal consistent goal-conditioned distortions, showing that current models remain sensitive to incentives and framing and lack robust safeguards against selectively misleading communication. We publicly release our corpus and code. https://github.com/PolydorosG/JANUS

cs.CL · cs.AI

新论文 · v1arXiv:2610.39870

Optimal self-orthogonal binary linear codes

Sascha Kurz

原文摘要

We complete the determination of the minimum possible length of eight-dimensional binary self-orthogonal codes with given minimum distance. For dimension nine we give exact values and bounds for minimum distance at most $64$. By exploiting the geometric reformulation of linear codes as multisets of points, we fully redo the classification for dimension at most seven in a simplified and computer-free manner. As a by-product we prove the non-existence of $[93,9,44]_2$ codes and show several divisibility results for putative optimal nine-dimensional binary codes.

math.CO · cs.DM · cs.IT

新论文 · v1arXiv:2610.39868

Parametric Gaussian Noise Mixer (PGNM): Unified Synthesis of Camera Noise and Film Grain

Mingyang Song

原文摘要

Noise synthesis is commonly approached through physical models of image formation or neural networks trained to generate noisy images. We explore a middle ground: an explicit statistical model fitted directly from paired clean and noisy images. A normalized convolution correlates a Gaussian field across pixels and color channels, and a lookup maps each Gaussian value to a residual given the clean intensity. Because the field stays standard Gaussian, the two components are fitted separately, without reproducing the physical process or training a neural generator. With one formulation for camera noise from SIDD and simulated film grain from Newson, the model has the lowest average distribution and correlation errors on SIDD among the evaluated methods and remains competitive on film grain. A denoiser trained on its noise comes closest to one trained on real noise, and a 1080p frame renders in 10 ms on GPU.

cs.CV

版本更新 · v3arXiv:2605.12964

Asymmetric Flow Models

Hansheng Chen, Jan Ackermann, Minseo Kim, Gordon Wetzstein, Leonidas Guibas

原文摘要

Flow-based generation in high-dimensional pixel spaces is difficult because velocity prediction requires modeling high-dimensional noise, even when data has strong low-rank structure. We present Asymmetric Flow Modeling (AsymFlow), a rank-asymmetric velocity parameterization that restricts noise prediction to a low-rank subspace while keeping data prediction full-dimensional. From this asymmetric prediction, AsymFlow analytically recovers the full-dimensional velocity, leaving the network architecture and the standard flow-matching training and sampling procedures unchanged. On ImageNet 256$\times$256, AsymFlow achieves a leading 1.57 FID, outperforming prior DiT/JiT-like pixel diffusion models by a large margin. AsymFlow also provides the first-ever route for finetuning pretrained latent flow models into pixel-space models: aligning the low-rank pixel subspace to the latent space gives a seamless initialization that preserves the latent model's high-level semantics and structure, so finetuning mainly improves low-level mismatches rather than relearning pixel generation. We show that the pixel AsymFlow model finetuned from FLUX.2 klein 9B establishes a new state of the art for pixel-space text-to-image generation, beating its latent base on HPSv3, DPG-Bench, and GenEval while qualitatively showing substantially improved visual realism.

cs.CV

版本更新 · v2arXiv:2610.02377

Flow Matching for Fast Posterior Sampling in Bayesian Inverse Problems

Jan Blechschmidt, Oliver G. Ernst, Moritz Poguntke, Björn Sprungk

原文摘要

Sampling from the posterior is the central task of computational Bayesian inverse problems. The standard workhorse in Bayesian inference---Markov chain Monte Carlo (MCMC)---is sequential, yields correlated samples, and must be rerun for each new observation. Conditional flow matching offers an amortized alternative: a transport map, trained once on joint samples of parameter and data, that yields independent approximate posterior samples for any observation at negligible online cost, without new likelihood evaluations. We give a careful, MCMC-literate assessment of flow matching for PDE-based inverse problems with function-valued parameters. Exploiting the flow's tractable density, we derive computable accuracy estimates for the learned posterior in total variation distance and Kullback--Leibler divergence and, moreover, propose a hybrid sampler that is made asymptotically exact by Metropolization. We validate the accuracy estimates and demonstrate amortization in several numerical examples, including electrical impedance tomography and a likelihood-free Lotka--Volterra model. Flow matching reproduces posterior means and credible bands closely at a fraction of the forward solves required by MCMC, but its accuracy degrades for concentrated posteriors and for observations that are atypical under the prior.

math.NA · cs.LG · stat.CO

新论文 · v1arXiv:2610.39863

From Trees to Binary Trees: Approximation Algorithms and a PTAS

Fabian Weisbrodt, Florian Werner

原文摘要

Kellerhals, Krebs, Nichterlein, and Schmid introduced the optimization problem Tree in Binary Tree (TBT), where, given a tree $T$, the task is to find a binary tree $S^*$ with $V(T)=V(S^*)$ that minimizes $\sum_{uv\in E(T)} \operatorname{dist}_{S^*}(u,v)$. They also introduced the variant Tree in Binary Tree With Steiner Nodes (TBT-SN), where the condition $V(T)=V(S^*)$ is relaxed to $V(T)\subseteq V(S^*)$. We give a simple linear-time $\frac{8}{3}$-approximation algorithm for TBT and a polynomial-time approximation scheme (PTAS) for TBT-SN. Our main building idea in both algorithms is to iteratively consider subtrees of the tree $T$ which contain all vertices below some vertex $v$ and construct a solution where $v$ has degree 1. For the PTAS, we cut $T$ into smaller trees, solve these instances, modify the obtained solutions, and combine them to achieve a solution for the original instance.

cs.DS · math.CO

版本更新 · v2arXiv:2505.07852

Joint Detection of Fraud and Concept Drift inOnline Conversations with LLM-Assisted Judgment

Ali Şenol, Garima Agrawal, Huan Liu

原文摘要

Detecting fake interactions in digital communication platforms remains a challenging and insufficiently addressed problem. These interactions may appear as harmless spam or escalate into sophisticated scam attempts, making it difficult to flag malicious intent early. Traditional detection methods often rely on static anomaly detection techniques that fail to adapt to dynamic conversational shifts. One key limitation is the misinterpretation of benign topic transitions referred to as concept drift as fraudulent behavior, leading to either false alarms or missed threats. We propose a two stage detection framework that first identifies suspicious conversations using a tailored ensemble classification model. To improve the reliability of detection, we incorporate a concept drift analysis step using a One Class Drift Detector (OCDD) to isolate conversational shifts within flagged dialogues. When drift is detected, a large language model (LLM) assesses whether the shift indicates fraudulent manipulation or a legitimate topic change. In cases where no drift is found, the behavior is inferred to be spam like. We validate our framework using a dataset of social engineering chat scenarios and demonstrate its practical advantages in improving both accuracy and interpretability for real time fraud detection. To contextualize the trade offs, we compare our modular approach against a Dual LLM baseline that performs detection and judgment using different language models.

cs.CL · cs.AI · cs.LG

新论文 · v1arXiv:2610.39857

Open-set multi-view seahorse re-identification from underwater photograph sequences

Tristan Fauvel, Sylvie Louisy, Patrick Louisy, Thomas Fauvel

原文摘要

Automated animal re-identification from photograph sequences can facilitate minimally invasive Capture-Mark-Recapture studies at scale. Yet, photographs taken during an observation session can vary greatly in quality, not always providing the necessary information for accurate re-identification. In this paper, we present new methods for taxonomic classification, sex determination, and open-set re-identification from image sequences. We train and evaluate our models on a dataset of 35,914 photographs of 474 individual seahorses observed in the wild. We demonstrate that our approach, by combining information from multiple photographs, yields state-of-the-art performance.

cs.CV

版本更新 · v2arXiv:2602.02100

The Verification Crisis: Expert Perceptions of GenAI Disinformation and the Case for Reproducible Provenance

Alexander Loth, Martin Kappes, Marc-Oliver Pahl

原文摘要

The growth of Generative Artificial Intelligence (GenAI) has shifted disinformation production from manual fabrication to automated, large-scale manipulation. This article presents findings from the first wave of a longitudinal expert perception survey (N=21) involving AI researchers, policymakers, and disinformation specialists. It examines the perceived severity of multimodal threats -- text, image, audio, and video -- and evaluates current mitigation strategies. Results indicate that while deepfake video presents immediate "shock" value, large-scale text generation poses a systemic risk of "epistemic fragmentation" and "synthetic consensus," particularly in the political domain. The survey reveals skepticism about technical detection tools, with experts favoring provenance standards and regulatory frameworks despite implementation barriers. GenAI disinformation research requires reproducible methods. The current challenge is measurement: without standardized benchmarks and reproducibility checklists, tracking or countering synthetic media remains difficult. We propose treating information integrity as an infrastructure with rigor in data provenance and methodological reproducibility.

cs.CY · cs.AI · cs.SI

新论文 · v1arXiv:2610.39854

Attention, Dynamic Wave and Hybrid Sequence Mixers in Language Models

Dmytro Babarytskyi, Oleksandr Babarytskyi

原文摘要

Large Language Models (LLMs) rely on Multi-Head Attention (MHA) for sequence modeling, but scaling to long contexts is constrained by the quadratic cost of attention, motivating sub-quadratic recurrent alternatives. This paper formalizes a Dynamic Wave recurrent layer, based on input-dependent exponential damping and cosine phase modulation, and evaluates three 10.94M-parameter configurations (Pure Transformer, Pure Wave, and Hybrid Wave-Attention) on TinyStories under perplexity, latency, throughput and peak VRAM profiling up to 4096 tokens. The Hybrid configuration reaches the best perplexity, 5.95 on the full held-out split against 6.93 for the Transformer and 7.26 for Pure Wave. The efficiency picture, however, is governed by implementation rather than by architecture. Against an unfused attention baseline, Pure Wave uses 78.7% less peak VRAM at a 4096-token context (1404.3 MB vs. 6601.6 MB); against a fused scaled-dot-product attention kernel the same comparison yields no advantage at all (1404.3 MB vs. 1403.6 MB), and in training mode Pure Wave is 8.3% more expensive than the Transformer. Likewise, hoisting elementwise terms out of the recurrence loop speeds the layer up 2.53-fold with no custom kernel, reducing its latency gap to attention from 2.48-fold to 1.12-fold. A hybrid stack is found to inherit the cost of its more expensive mixer rather than the advantage of its cheaper one, its peak VRAM matching the Pure Transformer to within 0.1%. A closed-form memory model reproduces these measurements and predicts the context length beyond which attention overtakes the vocabulary projection. The findings indicate that reported efficiency gains for recurrent attention replacements must be stated relative to a named attention backend and a named implementation, and that hybridization is best justified on quality grounds.

cs.LG · cs.CL

版本更新 · v2arXiv:2605.18346

Focused Forcing: Content-Aware Per-Frame KV Selection for Efficient Autoregressive Video Diffusion

Peiliang Cai, Evelyn Zhang, Jiacheng Liu, Hao Lin, Ruiqi Zhang, Weile Mo, Yue Ma, Shikang Zheng 等 11 位作者

原文摘要

Recent advances in autoregressive video diffusion have enabled sequential and streaming video generation. However, long-horizon generation requires increasingly large KV caches, making efficient compression without sacrificing quality challenging. Existing methods mostly select historical frames based on attention scores, but their context decisions remain coarse. When multiple frames are generated in the same chunk, these methods often apply a shared history selection to the whole chunk, score historical frames solely by attention, and assign head-wise budgets either uniformly or by attention-pattern heuristics rather than explicit head-importance estimation. We show that frames within the same generated chunk can depend on distinct historical frames, that the same historical frame can receive different attention scores as its relative temporal distance to the current frames changes, and that masking different heads induces unequal generation degradation. Motivated by these findings, we propose Focused Forcing, a training-free KV selection method that focuses cached history along both generated-frame and head dimensions. For each generated frame, Focused Forcing preserves the most relevant and distinctive historical frames by combining attention scores with diversity scores of historical frames, while assigning larger budgets to heads with higher estimated importance. Across multiple autoregressive generation paradigms, Focused Forcing achieves up to 1.48x end-to-end acceleration without training, while improving visual quality and text alignment. Our code is available at https://github.com/caipeiliang2004/Focused-Forcing.

cs.CV · cs.AI

版本更新 · v2arXiv:2610.34467

DynaFlow: Learning Free-Form Multi-Agent Orchestration

Ruhao Liu, Qi Li, Shizun Wang, Xinchao Wang

原文摘要

Multi-agent systems often rely on predefined collaboration structures, limiting their ability to adapt to diverse task requirements. We introduce DynaFlow, a framework that integrates free-form executable workflow synthesis into the action space of the main agent. The agent decides when collaboration is needed and generates task-specific programs that coordinate sub-agents through sequential or parallel execution, structured information exchange, and verification. Execution results and traces inform subsequent actions, enabling the agent to adapt collaboration as reasoning progresses. We train only the main agent through supervised fine-tuning on distilled trajectories, followed by refinement on examples filtered for unsupported premises, biased verification, and information-flow defects. All sub-agents remain fixed, separating orchestration learning from executor capability improvement. Across six benchmarks spanning mathematical reasoning, scientific question answering, and multi-hop retrieval, DynaFlow achieves 88.08% overall accuracy in Forced mode, outperforming the strongest evaluated baseline by 10.90 percentage points with the same Qwen3.5-27B backbone. Workflow analysis finds more explicit use, verification, and refinement of upstream outputs after training, highlighting cross-stage information flow as a key dimension of multi-agent orchestration.

cs.MA

新论文 · v1arXiv:2610.39853

KV Cache Quantization for Diffusion Language Models

Yan Tai, Yunan Ding, Guangzhao Li, Siqi Luo, Chaoyang Zhang, Guangtao Zhai, Guodong Guo, Xiaohong Liu

原文摘要

KV cache quantization is well studied for autoregressive language models, where cached prefix states are exact under causal attention and quantization is mainly a cache reconstruction problem. Diffusion language models (dLLMs) break this assumption. Fast dLLM inference reuses KV states across denoising steps, but bidirectional token representations can change as masked tokens are revealed, so the reused cache may already differ from full recomputation before quantization is applied. We study KV cache quantization under this cache-reuse error. We adapt KIVI and KVQuant to Fast-dLLM-style prefix and dual cache reuse, and introduce TRACE and dLLM-KVQ. TRACE reconstructs feasible dLLM reveal trajectories from prompt-answer pairs, providing calibration states that match iterative masked decoding. dLLM-KVQ uses these trajectories to choose per-layer K-cache entries that remain in high precision according to output- and attention-level behavior metrics, while applying standard low-bit value quantization. Across LLaDA and Dream evaluations on GSM8K, HumanEval, MATH, and MBPP, dLLM-KVQ achieves the highest average score among quantized methods across the reported 4-bit and 2-bit prefix/dual cache settings.

cs.CL

新论文 · v1arXiv:2610.39852

DWISE: Denoising Workload-Informed Scheduling Engine

Lie Li, Wen Li, Junxiao Shen, Jindong Gu, Guosheng Hu

原文摘要

Unlike autoregressive large language models (AR-LLMs), diffusion LLMs (dLLMs) generate text by iteratively denoising many tokens in parallel, offering a promising path to efficient inference. Existing serving frameworks, however, schedule dLLM requests with policies designed for AR models, namely first-come-first-served (FCFS) admission and prefill priority. We find that both policies are ill-suited to dLLMs, wasting much of their speed. FCFS ignores that a dLLM request's decoding cost is highly heterogeneous and unknown at arrival, causing severe head-of-line blocking. Prefill priority, widely used in AR-LLM serving to keep time-to-first-token low, sharply cuts dLLM efficiency, as rounds that run only prompts leave most of the batch idle. We present DWISE, a Denoising Workload-Informed Scheduling Engine for efficient dLLM serving. DWISE schedules requests by their remaining work, measured in a block-level unit shared by prefill and denoising, so both phases share a round. To predict the unknown denoising workload, DWISE combines a lightweight prompt-based predictor at admission with a single refinement from the first denoising round's hidden states, which capture model-specific information absent from the prompt and require no additional model execution. Implemented in SGLang and evaluated on five models from three dLLM families under bursty workloads, DWISE reduces mean end-to-end latency over SGLang by 5.4$\times$ on average and up to 14.9$\times$ while preserving task accuracy. The code will be released publicly.

cs.LG

新论文 · v1arXiv:2610.39850

The information geometry of the Sinkhorn divergence on a Hilbert manifold

Paul Navas

原文摘要

The Sinkhorn divergence, the debiased entropic optimal transport cost, determines a metric tensor on probability measures defined along curves, due to Lavenant et al. We construct a real-analytic Hilbert manifold of measures, the smooth stratum, on which this tensor is a weak Riemannian metric, and we continue Eguchi's construction of a geometry from a divergence to the third order. The stratum consists of the measures with a positive Sobolev density on a closed Riemannian manifold, and the strata form a scale whose intersection is the Fréchet manifold of smooth positive densities of Ay. Following it, we characterize the metric through the gradients of mean values: its cotangent form is a spectral reweighting of the Dirichlet form of a Markov chain, so the metric is of the same kind as the Fisher-Rao and the Otto metric, and on the stratum it is bounded by a multiple of the Fisher-Rao metric (but not conversely). At third order, the Sinkhorn divergence is self-dual: both connection forms of Eguchi's construction coincide with the Levi-Civita form of the metric, and the cubic tensor vanishes, as a consequence of the transposition symmetry of the transport problem. On every stratum, the Levi-Civita form is represented, so the stratum has a real-analytic Levi-Civita connection, which on the Fréchet manifold is the connection of Ay and Schwachhöfer.

math.ST · cs.IT

新论文 · v1arXiv:2610.39849

Integrating Perception, Comprehension, and Projection for Alert Validation: A Unified Cyber Situational Awareness Framework

Navodika Karunasingha, Mohan Baruwal Chhetri, Surya Nepal, Cecile Paris, Salil S. Kanhere

原文摘要

Security Operation Centres (SOCs) must validate large volumes of alerts with limited analyst resources, yet alerts often lack sufficient context to determine whether suspicious activity is attack related. While provenance-based intrusion detection research has progressed from perception-level signals to comprehension- and projection-level cyber situational awareness (CSA), it remains unclear whether richer CSA consistently improves alert validation. This paper presents the first systematic evaluation of perception-, comprehension-,and projection-level CSA within a unified alert validation framework. Using Captain and Kairos, we generate multi-level CSA and formulate alert validation as a positive-unlabelled multi-instance learning (PU-MIL) task over CSA-derived behavioural hypotheses. Experimental evaluation on the DARPA TC dataset shows that comprehension-level CSA substantially improves validation by incorporating causal attack context. Relative to the Queue and Max. CSG baselines, PU-MIL achieves up to 28.6 and 75 percentage-point gains in Recall@5, and 24.6% and 86.8% reductions in mean positive rank on Cadets and Theia, respectively. Projection-level CSA can yield additional benefits, reducing the mean positive rank on Theia from 22.0 to 8.5 at a 40-minute projection horizon, although its effectiveness varies across projection horizons. Overall, richer CSA does not necessarily lead to better alert validation. Rather, its value depends on when and how it is introduced, motivating adaptive CSA progression that selectively enriches alerts according to validation requirements.

cs.CR

版本更新 · v2arXiv:2604.22742

Boolean PCSPs through the lens of Fourier Analysis

Demian Banakh, Katzper Michno

原文摘要

We develop an analytical framework for Boolean promise constraint satisfaction problems (PCSPs) that studies polymorphisms through the notion of influence from Fourier analysis of Boolean functions. Extending the work of Brakensiek, Guruswami, and Sandeep [ICALP'21] on Ordered PCSPs, we identify two general phenomena in Boolean minions indicative of hardness or tractability: (1) preservation of coordinate influence under random 2-to-1 minors and (2) the presence of sharp thresholds. We demonstrate that these phenomena occur in broader settings than previously established, yielding new hardness/tractability results for minions consisting of unate or polynomial threshold functions.

cs.CC

新论文 · v1arXiv:2610.39848

LiDAR-SHIELD: A Fully Dockerized Adversarial Robustness Assessment Framework for LiDAR-Based 3D Perception Models

Alexandros Gkillas, Ioulia Kapsali, Christos Anagnostopoulos, Aris S. Lalos

原文摘要

LiDAR-based 3D perception underpins autonomous driving and robotics, yet deploying these models safely requires stress-testing them against adversarial inputs before real-world use. Although state-of-the-art adversarial attacks exist, their implementations are often fragmented across individual repositories, each tied to a specific architecture or task and requiring manual environment reconstruction and model-specific integration code. We present LIDARSHIELD, a unified and fully Dockerized framework that reverses this workflow: rather than installing the target model inside the toolkit, LIDARSHIELD exposes any model through three lightweight HTTP endpoints, namely /info, /forward, and /gradient As a result, connecting a new model or dataset requires only a single loader function or adapter. The framework supports both 3D semantic segmentation and 3D object detection through the same engine-service protocol, integrating a task-aware suite of state-of-the-art attacks, including PGD and Square Attack for segmentation, and point perturbation, point removal, point injection, and LiDAttack for detection. This enables systematic exposure of corner cases and vulnerabilities across heterogeneous LiDAR backbones. The LiDAR-Shield codebase is publicly available at https://github.com/alexandrosgk/LiDAR-shield.

cs.CV · cs.AI

新论文 · v1arXiv:2610.39846

TokWatt: an open, boundary-annotated database of AI inference energy, from chat answers to agents

Abderaouf Bahi

原文摘要

Electricity supply is becoming a binding constraint on the deployment of artificial intelligence (AI), yet public evidence on the energy used by a single AI task is scattered and hard to compare. Published figures for one chatbot answer differ by two orders of magnitude, largely because studies count different parts of the system and measure different units of work. We present TokWatt, an open database of 134 published estimates of AI inference energy drawn from 13 sources (2023-2026) and covering 20 tasks, from classic machine-learning inferences to chat answers, reasoning answers, long-document queries, image and video generation, and coding agents. Each estimate records its functional unit, five measurement-boundary components, method, reported interval, and the location of the value in its source. A reproducible procedure scales estimates to a common server-plus-facility boundary and derives per-task reference values with ranges. The reference values span six orders of magnitude: 0.34 Wh for a short chat answer (range 0.16-0.99 Wh), 4.7 Wh for a reasoning answer, 124 Wh for a short video clip, and 150 Wh for one task delegated to a coding agent. Five independent estimates for default production chat systems agree within a factor of 1.8, and the reference values shift by at most 12% under alternative harmonization parameters. A simple identity links personal use to grid load: E watt-hours per person per day across one billion people corresponds to an average draw of E/24 gigawatts. Data and code are released at https://github.com/AraoufBh/tokwatt, and an online calculator at https://araoufbh.github.io/tokwatt/.

cs.CY

版本更新 · v2arXiv:2604.13151

Exploration and Exploitation Errors Are Measurable for Language Model Agents

Jaden Park, Jungtaek Kim, Jongwon Jeong, Robert D. Nowak, Kangwook Lee, Yong Jae Lee

原文摘要

Language Model (LM) agents are increasingly used in complex decision-making tasks, from AI coding to physical AI, where they must both explore a search space and exploit acquired knowledge effectively. Although exploration and exploitation play different roles in decision-making, it remains challenging to systematically distinguish exploration and exploitation without access to the agent's internal policy, and to determine when either constitutes an error. To address this, we design partially observable 2D grid maps paired with an unknown symbolic task directed acyclic graph (DAG), abstracting practical agentic AI scenarios. We then propose a policy-agnostic metric that detects unproductive traversals, while permitting necessary backtracking. We attribute each error to exploration or exploitation based on whether the current state admits productive exploration, exploitation, or both. Using generated maps and task DAGs with controllable difficulty, we evaluate 35 frontier LM agents and find that even state-of-the-art models struggle to reach the goal within the step budget. $\log$ exploration and exploitation errors exhibit strong relationships with task success ($R^2 = 0.885$ and $0.875$, respectively), providing empirical evidence that the metric captures behavior relevant to task success. Beyond success rate, the metric reveals distinct exploration and exploitation behaviors among models with similar success rates. We further find that a compact memory harness achieves higher task success than a reasoning baseline that retains its own reasoning traces. Finally, we extend our metric to an iconic video game, Minecraft Clone, beyond our synthetic environment. We release our code at https://github.com/jjj-madison/measurable-explore-exploit.

cs.AI