跳到正文
研踪PaperTrail

YOUR RESEARCH, IN SIGHT

让下一篇重要论文，不再错过。

覆盖计算机全学科，以 AI 为快捷入口。关注方向、作者和关键词，把发现变成持续的研究积累。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+0新增论文
+0其中 AI 相关
+1已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-09 00:58 · 全站最近入库：2026-11-08 06:02各信源进度 →

统计口径与数据新鲜度

统计 2026-11-08 01:35 — 2026-11-09 01:35（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：2026-11-09 00:58最近发现新增版本：2026-11-08 06:02库内论文最新更新时间：2026-11-06 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。

17,100 已收录论文14,661 AI 相关4,057 有修订112 HF 社区发现

浏览论文库

研究日报 / 周报 →按信源浏览 →

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 17,100 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-09 00:56（北京时间）
新论文 · v1arXiv:2610.39658

Towards Audience-Driven Real-Time Data Storytelling

Angelica Lo Duca

原文摘要

This paper introduces the concept of Audience-Driven Real-Time Data Storytelling (ADRDS), a framework that utilizes participatory and social sensing to enable audiences to actively shape the presentation of data stories. At its core, ADRDS employs a pre-authored decision tree, where the presenter defines the data story through a branching three-act arc (setup, conflict, and resolution). Each node contains a question, branching rules, thresholds for coverage, consensus, and confidence intervals, as well as one or more potential calls to action (CTAs). Audience members connect to the ADRDS framework via a URL on their mobile devices and answer questions in real time. ADRDS evaluates responses against the decision tree in real-time, following strong branches when thresholds are satisfied or activating fallback paths when signals are inconclusive. ADRDS then selects the final action consistent with the traversed path and the audience's demonstrated readiness, subject to fairness safeguards. ADRDS transforms presentations of data stories into audience-shaped decision processes, where live data actively drives the narrative. This work paves the way towards a new form of data storytelling, where the final data story depends on the audience's background and interests.

cs.HC

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2605.11012

Inverse Design of Metainterfaces towards Static Friction Control: Beyond the Hertzian Limit

Jacopo Bilotto, Arnav Singhal, Joaquin Garcia-Suarez, Gaëtan Cortes, Lucas Fourel, Jean-François Molinari

原文摘要

Programming the static friction of mechanical interfaces is critical for soft robotics, haptics, and precision gripping. Static friction is governed by the real contact area, and standard rough surfaces exhibit a linear area-load scaling inherent to classical models, severely restricting their functional range. Here, we propose a framework for the inverse design of tribological metainterfaces engineered for programmable contact behaviors. By utilizing general axisymmetric asperities, we unlock nonlinear macroscopic responses unattainable by standard Hertzian contacts. To solve the inverse problem, we embed a fully differentiable contact mechanics engine within a neural network and a quadratic optimizer. We leverage regularized physical gradients to automatically discover non-standard topographies that reproduce complex target contact laws, with only a few asperities in unit cells. The predicted designs are validated against high-fidelity Boundary Element Method (BEM) simulations. By front-loading computational costs during training, the neural surrogate yields near-instantaneous design inference, bypassing the bottlenecks of traditional multi-start optimization. This framework bridges data-driven optimization and rigorous physics, offering a scale-invariant pathway for the high-throughput discovery and real-time control of functional tribological surfaces.

cond-mat.soft · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2510.09177

Distributionally robust approximation property of neural networks

Mihriban Ceylan, David J. Prömel

原文摘要

The universal approximation property uniformly with respect to weakly compact families of measures is established for several classes of neural networks. To that end, we prove that these neural networks are dense in Orlicz spaces, thereby extending classical universal approximation theorems even beyond the traditional $L^p$-setting. The covered classes of neural networks include widely used architectures like feedforward neural networks with non-polynomial activation functions, deep narrow networks with ReLU activation functions and functional input neural networks.

stat.ML · cs.LG · math.FA · math.PR

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39655

The Support-Distribution Conditions for MSRD Codes Reduce to a Single Inequality

Xiang Fan

原文摘要

Byrne, Gluesing-Luerssen, and Ravagnani determined the number of codewords with a prescribed nonzero support in a linear maximum sum-rank distance code whose blocks have a common number of columns. Their formula depends only on the code parameters and the support dimensions. Evaluating it for arbitrary parameters gives a family of necessary nonnegativity conditions for existence. We prove their conjecture that this family is equivalent to the condition at the vector of total dimension $d+1$ obtained by filling the longest blocks first, where $d$ is the distance parameter. The resulting explicit criterion applies over every finite field $\mathbb F_q$, for all row numbers $n_i\leqslant m$ and $1\leqslant d<N$, where $m$ is the common number of columns and $N$ is the total number of rows. It also covers the boundary cases in which this vector fills each of its nonzero blocks completely, not explicitly included in the conjecture's displayed parameter range. The proof uses a generating function and an exchange identity to reduce the problem to vectors obtained by filling the longest blocks first. We then estimate alternating sums of elementary symmetric polynomials, with a separate argument for $q=2$. The dual conditions admit the same reduction. For codes that exist, the joint condition follows from sphere-packing bounds for the code and its dual. An infinite family of parameters satisfies the joint condition but admits no such linear code.

cs.IT

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39653

On edge-ordered Ramsey numbers of hypergraphs

Martin Balko, Jelena Glišić

原文摘要

For an integer $h\geq2$, an edge-ordered $h$-uniform hypergraph $H^<$ is an $h$-uniform hypergraph with a linear ordering of its hyperedges. For an integer $q \geq 1$, the edge-ordered Ramsey number $R_{\mathrm{edge}}(H^<;q)$ of $H^<$ is the smallest positive integer $N$ such that there exists an edge-ordered $h$-uniform hypergraph $G^<$ on $N$ vertices such that, for every $q$-coloring of the hyperedges of $G^<$, there is a monochromatic copy of $H^<$ in $G^<$ preserving the hyperedge order. In this note, we show that, for every edge-ordered $h$-uniform hypergraph $F^<$ on $n$ vertices with $m\geq1$ hyperedges, $R_{\mathrm{edge}}(F^<;q) \leq R_h(F;q)^{Cm^2}$ where $F$ is the underlying unordered hypergraph, $R_h(F;q)$ is its standard $q$-color Ramsey number, and $C=C(h,q)>0$ is a constant. For fixed $h\geq3$ and $q\geq2$, this shows that prescribing the hyperedge order does not require an increase in the tower height of the classical upper bound. It also settles a problem posed by Fox and Li (2020), and for $h=2$ gives the bound $2^{O(n^4\log n)}$, improving their bounds for every fixed $q \geq 3$.

math.CO · cs.DM

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2407.10247

Strategic Integration of Artificial Intelligence in the C-Suite: The Role of the Chief AI Officer

Marc Schmitt

原文摘要

The integration of Artificial Intelligence (AI) into corporate strategy has become critical for organizations seeking to maintain competitive advantage in the digital age. Although organizations increasingly rely on AI as a strategic and organizational resource, existing C-suite roles remain only partially equipped to govern, integrate, and leverage it coherently at the enterprise level. Organizations vary in their responses. Some create a dedicated Chief AI Officer (CAIO), others extend existing mandates into hybrid roles, and still others coordinate AI through federated structures. This paper develops a role-design theory to explain this variation. I identify three properties that distinguish AI systems from earlier cross-cutting enterprise technologies, distributed accountability for judgment, upstream governance, and non-stationarity, and three configurations through which organizations respond: concentrated extension, distributed extension, and role creation. Four propositions specify when a dedicated CAIO emerges, what form an organization's response takes, when the role is effective, and how configurations evolve. Drawing on the first archival census of executive AI leadership, I find that the role has diffused sharply since 2023, that role creation and mandate extension are advancing in near parity, that adoption concentrates where digital leadership structures already exist, and that executive tenures contradict practitioner accounts of a transient role. The paper contributes to research on executive leadership, AI governance, and the design of AI-driven organizations.

cs.CY · cs.AI · cs.LG · econ.GN

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39650

Witnessed Boolean Contracts: Sound Local Distance Guarantees for Approximate Implicit CSG

Jeremy Pei

原文摘要

Hard min/max operations preserve the sign of exact implicit solids but generally destroy signed-distance semantics. Arbitrarily small, sign-preserving value perturbations can rotate the normalized gradient by nearly a right angle and displace a gradient-projected boundary witness by a finite amount. We formalize this as the absence of a value-only stability modulus for witness projection. We introduce a witnessed Boolean contract checking value and gradient errors, positive reach, active-branch separation, and a Lipschitz visibility margin at the approximate witness. A sharp normalization lemma bounds the normal-direction error. An enlarged neighborhood permits transfer of the inactive operand's bound between the approximate witness and the true foot point. Under the stated analytic and domain assumptions, passing the checks identifies a true point on the union boundary and bounds distance deficiency by the active operand's pointwise value error. Failed or unavailable checks return Unknown. The argument extends to finite hard unions by testing every inactive operand. An executable proof-chain audit finds no violations on 75,738 certified samples. A broader analytic stress test covers spheres, capsules, ring tori, three Boolean operators, and a four-leaf union. Of 96,000 narrow-band samples, 87,469 pass with zero oracle violations; per-case coverage ranges from 72.67% to 98.30%. Removing visibility admits 75 truly invalid witnesses, showing that branch separation alone is insufficient. Non-union cases remain empirical tests, not new theorems.

cs.CG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39649

Look at Where You Think: Localization-Head Alignment for Fine-Grained Perception in Thinking VLMs

Yuan Wang, Wei Wu, Hongning Wang

原文摘要

Thinking vision-language models reason for thousands of tokens, yet still miss small details in high-resolution images. Zooming methods pay for extra inference passes, and outcome rewards never tell the model where to look at. In Qwen3.5-4B, we find that the bottleneck is selection, not seeing: a few attention heads, which we call localization heads, pick the evidence out of thousands of visual tokens; ablating four of them costs 13.2 points more than ablating four random heads, and their focus fades as the reasoning grows. We therefore propose Localization-Head Alignment (LHA), which makes attention a training signal: whenever the reasoning states a visual fact, these heads learn to look at the focal region rather than the background, as an auxiliary loss in both SFT and RL training. On benchmarks of V*, HR-Bench 4K and TreeBench, LHA lifts Qwen3.5-4B from 74.1% to 76.6% accuracy, the best single-pass 4B result we evaluate, ahead of multi-pass zooming and released 7B models. Alignment alone adds 1.8 points over otherwise identical training while shortening reasoning by 20%, and the gains carry over to Qwen2.5-VL-3B.

cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39648

Annotation-Free Representation Learning for Polyp Retrieval in Colonoscopy Videos

Luca Parolari, Lamberto Ballan, Carlo Biffi, Loic Le Folgoc

原文摘要

Colonoscopy is inherently comparative: reliable visual assessment of polyps often depends on complementary views acquired under different viewpoints and imaging conditions. Retrieving such views requires polyp tracklet representations, which recent contrastive methods learn from temporal cues but still build from manually annotated bounding boxes. We propose an annotation-free framework that constructs tracklet views directly from raw colonoscopy videos, enabling temporal self-supervised learning on full-procedure videos, including frames without visible polyps. On REAL-Colon, our method achieves state-of-the-art results on tracklet retrieval and re-identification, outperforming both self-supervised and supervised baselines. We then introduce few-shot polyp retrieval, a task closer to clinical use, where a few observations of a target polyp are used to retrieve its remaining tracklets: with only two examples, our method surpasses most competing methods given ten. Finally, we show that a modest number of unlabeled procedures suffices for competitive performance, making on-site model development more practical. Code is available at https://github.com/lparolari/freerep.

cs.CV

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.24616

From Tokens to Structure: Investigating Visual Attention in Progressive Code Rendering

Xiaotian Su, Jan Brasser, David Robert Reich, Lena A. Jäger, April Yi Wang

原文摘要

AI coding interfaces present generated code either all at once or token-by-token. These rendering strategies reflect model generation rather than how programmers actually read code: selectively, non-linearly, and guided by the structure. We argue that code rendering is a first-class interaction primitive that shapes how programmers read and understand code. To explore this design space, we introduce structured rendering, a technique that reveals code in semantically meaningful chunks derived from its syntactic hierarchy, exposing high-level structure before low-level details. To isolate rendering effects on visual attention, we conducted an eye-tracking study with 53 participants comparing static, character-based, and structured rendering. Our findings show that rendering alters visual attention and reading behavior: dynamic rendering induces fewer but longer fixations and more sustained focus, while structured rendering further guides attention toward semantically meaningful units and supports high-level understanding. We release the anonymized dataset with an interactive demo (https://codegaze.vercel.app/) to support future research.

cs.HC

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39646

FlexTry-On: A Time-Efficient Framework for Multi-Scenario Virtual Try-on

Phuong Dam, Quyen Nguyen Duc, Son Hoang Minh, Daeyoung Kim

原文摘要

Virtual try-on has achieved remarkable progress in recent years, yet most existing methods are designed for a specific scenario, such as cloth-to-person or person-to-person try-on, and often require separate conditioning strategies and processing pipelines for different settings. We present FlexTry-On, a unified and time-efficient diffusion-based framework for diverse virtual try-on scenarios, including cloth-to-person, person-to-person, cross-domain, and in-the-wild settings. FlexTry-On adopts a common conditioning formulation, where a text-aware condition extraction strategy converts heterogeneous garment sources into a unified representation, and a dual-conditioning diffusion transformer jointly models garment appearance and person identity. Extensive experiments demonstrate competitive performance on both cloth-to-person and person-to-person benchmarks while maintaining fast inference. Moreover, the same framework can be directly applied to unseen cross-domain scenarios without requiring separate architectures or scenario-specific processing pipelines. These results highlight the potential of unified and efficient virtual try-on frameworks as a practical step toward more flexible and widely applicable virtual try-on systems.

cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39645

Carbon-Aware Electric Vehicle Routing Problem with Decision-Focused Learning

Jaume Reixach, Wanru Chen, Jiaming Wu, Balázs Kulcsár

原文摘要

Electric vehicles produce no tailpipe emissions, but charging them produces emissions that depend on the time-varying generation mix of the electricity consumed. We study an Electric Vehicle Routing Problem with a carbon-aware objective that accounts for charging-related emissions alongside conventional routing costs. The problem determines routing, charging, and waiting decisions, allowing charging demand to be shifted toward periods with lower grid carbon intensity. Because these decisions depend on forecasts of future carbon intensity, we propose a decision-focused learning approach so that the training of the model employed for predicting the carbon intensity of the grid over time is solver-aware. In particular, we train a linear predictor using approximate SPO+ directions and employ a time-limited Large Neighborhood Search followed by exact postprocessing as the underlying solver. Experiments on realistic Germany-based instances show that explicitly considering charging-related emissions reduces realized emissions by 28.76% on average relative to carbon-blind optimization, although at the expense of higher non-carbon operating cost. Moreover, decision-focused training reduces realized total cost by 2.30% on average relative to conventional prediction-focused learning and performs better on 75.5% of the test instances, despite substantially lower pointwise predictive accuracy. Finally, we show that its average advantage remains positive across varied solver configurations and carbon-cost weights.

math.OC · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39641

Rebuilding a Field's Open-Problem Catalog from the Literature: A Two-Layer Knowledge Graph, and Why Status Must Stay Proposed

Madhulatha Mandarapu, Sandeep Kunkunuru

原文摘要

A research field records its open problems in surveys, keynotes and community reports. Those are prose, written every few years, and nobody can ask them who works on a problem now or whether it was settled last year. We argue the problems should be data, and we build the system that makes them so: a hand-curated problem layer carried in unchanged, under an evidence layer rebuilt from a bibliographic index, joined by typed, provenance-carrying links in one graph database and queryable from one language. A language model does four jobs -- per-paper extraction, catalog grounding, per-paper status verdict, and idea generation -- each constrained to a candidate set so out-of-set answers are refused rather than stored. We report a study on two curated catalogs, database research (1,053 problems) and AI research (1,000 problems), each with a prior catalog-only graph. The rebuilt graph answers every question the catalog-only graph answered -- 47 of 47 on databases, 34 of 35 on AI -- plus questions the catalog-only graph refused: recent workers, untouched problems, methods tried, and outside papers bearing on a problem (336 problems, 770 works on databases). The capability we most wanted does not survive its own audit. Of 41 solves and refutes verdicts hand-checked against the papers, 9 are correct, 30 partly correct and 2 wrong; no refutes verdict (0 of 34) disproves an open problem, and only 1 of 7 solves closes the central question. We report that failure in full and draw the design conclusion: a status change must be proposed with its evidence and never applied, making the human reviewer part of the architecture. We close with the data-engineering problems this raises: entity resolution across curated and harvested layers, verifying model status claims, freshness under source quotas, and evaluation without ground truth.

cs.DB · cs.AI · cs.DL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39640

Early Prediction of Cross-Service Implementation Scope: Evidence and Limits from Multiple Microservice Projects

Zhoumaocuo, Xiaozhou Li

原文摘要

Microservice architectures aim to localize change, yet feature requests may require implementation across service boundaries. We investigate how well such downstream scope can be anticipated before implementation begins. Using issue--pull-request--commit traceability and implementation evidence corrected through targeted human adjudication, we reconstruct 866 repository-traceable feature requests from 13 open-source microservice projects and map their implementations to validated service boundaries. Cross-service changes occur in 101 requests (11.7%) and exhibit broader implementation footprints, particularly higher code churn, but no clear difference in lead time. Pre-implementation evidence provides limited cross-project predictive signal: the strongest leave-one-system-out model reaches a project-macro Macro-F1 of 0.484, but its advantage over the majority baseline is not significant after Holm correction and is sensitive to small evaluation folds. A local LLM offers no reliable improvement over classical models. Downstream workload is substantially harder to estimate, as learned models do not consistently outperform training-fold median baselines. A simple service-mention heuristic also performs comparably to learned scope triage. These findings show that early feature-request evidence is more useful for lightweight cross-service scope triage than for automated scope determination or precise workload estimation.

cs.SE

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2605.09383

Safety-Critical LiDAR-Inertial Odometry with On-Manifold Deterministic Protection Level

Yueqi Zhu, Yan Pan, Chufan Rui, Jiasheng Luo, Bo Zhou

原文摘要

In safety-critical scenarios, the protection level of the autonomous navigation system is crucial for enabling mobile robots to perform safe tasks. However, existing studies on probabilistic navigation systems for robots usually perform offline accuracy evaluations using limited datasets and assume that the results can be applied to unknown real-world environments. As a result, current autonomous mobile robots often lack protection levels for online safety assessment. To fill this gap, we propose a safety-critical LiDAR-inertial odometry (LIO) that provides deterministic protection levels based on a novel on-manifold set-membership filter. By adopting the unknown but bounded assumption, we derive a neat closed-form relationship between point cloud noise and the uncertainty of the estimation from the iterative closest point algorithm. Using this relationship, we design an on-manifold ellipsoidal set-membership filter (OM-ESMF) and implement it within the LIO system. Leveraging the properties of the set-membership filter, our system offers the feasible sets of the estimated locations as the deterministic protection levels, serving as safety references for the robots' downstream autonomous operations. The experimental results show that our system can provide effective deterministic online safety references for diverse robots in various environments.

cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2512.10753

Quantifying displacement: an urban expansion consequence via persistent homology

Rita Rodríguez Vázquez, Manuel Cuerno

原文摘要

Population displacement is a housing-related involuntary residential dislocation. It has become increasingly widespread in many cities, particularly in neighbourhoods undergoing rapid economic and demographic change, and measuring it is essential to assess the social consequences of urban transformation and housing market pressures. Despite its relevance, quantifying displacement presents difficulties due to limited replicability across cities and time periods and the need to analyse long time spans: displacement is a gradual process, impossible to capture in one data snapshot. We introduce a novel tool to overcome these difficulties. Using publicly available address change data, we construct four cubical complexes simultaneously incorporating geographical and temporal information of people moving, and analyse using Topological Data Analysis tools. Finally, we demonstrate this method through a 20-year case study in Madrid, Spain. The results demonstrate its ability to reveal patterns consistent with population displacement and identify the neighbourhoods and years potentially affected--patterns not observable from yearly maps of the raw address change data.

cs.CG · cs.SI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39639

Where Do Vision Language Models Fail at 3D Reasoning?

Khai Le-Duc, Nguyen Kim Hai Bui, An Ngo, Nam D. Hoang, Yueqi Song, Jiarui Liu, Thai Quoc Hoang, Fang Wu 等 14 位作者

原文摘要

Vision-language models can get individual judgments right while struggling to recover and consistently use the exact properties of a 3D object. To characterize this gap, we introduce BedBench, a controlled benchmark of thirteen procedurally generated tasks spanning individual judgments and composite answers. Across twelve open-weight and four frontier models, per-color counting exposes a pronounced weakness: humans achieve 93.3% accuracy, compared with 7.2% for the best model. Targeted fine-tuning raises one model's accuracy from 3.0% to 68.6%, but reveals a further difficulty: its answers about the most common colors remain inconsistent with its own per-color counts. Supplying those counts as text improves its judgments, showing that recovering a property does not mean using it consistently. Reasoning, scaling, and added visual cues each improve individual judgments, but none makes exact answers reliable. To locate the difficulty, we probe Qwen3.5-27B: exact per-color counts remain recoverable at the language model's input, yet the model's answers rarely recover them (65.0% for the probe vs. 3.0% for the model). Image-removal controls further show that some apparent successes require no visual evidence. These findings motivate evaluating 3D reasoning by exact correctness, consistent use, and image dependence, rather than accuracy alone. Code and dataset are publicly available: https://github.com/leduckhai/BedBench

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39638

MRUD: Maximally Redundant Unsigned Digit Number Systems for High-Performance Arithmetic Units

Samaneh Rahimi, Ali Azarpeyvand

原文摘要

Signed digit redundant number systems such as Maximally Redundant Signed Digit (MRSD) enable carry-free addition but introduce significant overhead in multiplication due to the presence of both positive and negative bits. While Unsigned Digit Redundant (UDR) systems have shown promise in radix-2, their extension to higher radices remains unexplored. This work presents the first generalized Maximally Redundant Unsigned Digit (MRUD) number systems for high-radix arithmetic, extending UDR representations from radix-2 to radix-$2^h$ while preserving carry-free addition. The proposed framework includes digit encoding, digit extension, arithmetic operations (addition, subtraction, and multiplication), sign inversion, forward/reverse conversion algorithms, overflow/underflow detection, and hardware architectures for both signed and unsigned MRUD representations. Unlike MRSD, which is inherently limited to signed arithmetic, MRUD supports both signed and unsigned representations, similar to conventional binary systems. Evaluations across three multiplier modes---RRR (Redundant inputs, Redundant output), BBR (Binary inputs, Redundant output), and BBB (Binary inputs, Binary output)---as well as Multiply-Accumulate (MAC) units, using the FreePDK 45nm library, demonstrate that MRUD achieves up to 30\% lower delay, 26\% smaller area, and 87\% lower power than MRSD. In BBR mode, MRUD achieves up to 43\% delay reduction over the best conventional architecture for 32$\times$32 operands, while in Multiply-Accumulate (MAC) units, performing reverse conversion only once at the end yields up to 64\% higher throughput than DesignWare and over 150\% higher than Wallace. These results demonstrate that MRUD enables high-radix redundant arithmetic with significantly improved multiplier and MAC performance, making it an efficient alternative to conventional signed digit redundant number systems.

cs.AR · eess.IV

查看论文
arXiv ↗PDF ↗
版本更新 · v6arXiv:2602.12851

Chimera: Neuro-Symbolic Attention Primitives for Trustworthy Dataplane Intelligence

Rong Fu, Xiaowen Ma, Kun Liu, Wangyu Wu, Ziyu Kong, Jia Yee Tan, Tailong Luo, Xianda Li 等 10 位作者

原文摘要

Deploying expressive learning models directly on programmable dataplanes promises line-rate, low-latency traffic analysis but remains hindered by strict hardware constraints and the need for predictable, auditable behavior. Chimera introduces a principled framework that maps attention-oriented neural computations and symbolic constraints onto dataplane primitives, enabling trustworthy inference within the match-action pipeline. Chimera combines a kernelized, linearized attention approximation with a two-layer key-selection hierarchy and a cascade fusion mechanism that enforces hard symbolic guarantees while preserving neural expressivity. The design includes a hardware-aware mapping protocol and a two-timescale update scheme that together permit stable, line-rate operation under realistic dataplane budgets. The paper presents the Chimera architecture, a hardware mapping strategy, and empirical evidence showing that neuro-symbolic attention primitives can achieve high-fidelity inference within the resource envelope of commodity programmable switches.

cs.NI · cs.AI · cs.CR · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v5arXiv:2607.19305

Riemannian Deep Learning: Modules, Networks, and Geometries

Ziheng Chen

原文摘要

Deep neural networks on manifold-valued representations have attracted growing interest, but many basic components remain tied to specific manifolds, rely on Euclidean approximations, or require costly and numerically fragile geometric operations. This thesis develops a unified framework for Riemannian deep learning from three complementary perspectives: reusable neural modules, manifold-specific network architectures, and the design of underlying geometries. It generalizes batch normalization from Euclidean spaces and individual manifolds to broad classes of Lie groups and gyrogroups, and extends multinomial logistic regression from Euclidean space to SPD manifolds and then to general Riemannian manifolds. It further develops neural networks for several important geometric representations, including an unconstrained model of hyperbolic space, Busemann-based hyperbolic learning, and full-rank correlation matrices. Finally, it introduces adaptive and computationally efficient Riemannian metrics on SPD manifolds, including learnable Log-Euclidean geometries and fast, stable Cholesky-based geometries. The proposed methods are supported by theoretical analysis and validated through numerical experiments and applications in vision, signal processing, graph learning, and genomics.

cs.LG · cs.AI · math.DG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39634

Evaluating Conversational Explainability in Healthcare: Comparing the Explainability Assistant and a Traditional XAI Dashboard

Eduard Barbu, Rodion Krjutškov, Anneli Uusküla

原文摘要

Conversational interfaces may make explainable AI (XAI) more accessible to users without machine learning expertise, but their benefits over dashboards remain insufficiently characterised in healthcare. Existing systems have typically been evaluated in isolation, leaving unclear whether conversational interaction improves interpretability beyond the underlying explanation methods. We report a within-subjects user study comparing the Explainability Assistant, an LLM function-calling conversational XAI system, with a traditional dashboard for explaining a heart disease classification model trained on the Cleveland Heart Disease dataset. Both interfaces expose the same Decision Tree classifier and post-hoc explanation methods: SHAP feature attribution, DiCE counterfactuals, and what-if analysis. This design supports comparison of interaction modalities while holding the model and explanation methods constant. Five participants with healthcare backgrounds, from medical students to practising clinicians and public health researchers, completed parallel task blocks on both interfaces; three participated in follow-up interviews. Task accuracy, Likert ratings, free-text questionnaire responses, and interviews revealed complementary strengths and weaknesses rather than a clear preference for either modality. The assistant offered easier entry and patient-specific explanations, while the dashboard supported simultaneous inspection of global and local information. All three interview participants recommended integrating conversational and dashboard elements within a single clinical workflow.

cs.HC

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2608.30685

ATLAS: Dual-Horizon Diagnostic Evaluation for Tool-Use Agents in Meituan Xiaotua

Wei Chen, Peilun Zhou, Zhaoyu Hu, Jiajun Chai, Derong Xu, Zhongni Hou, Yufei Zhang, Guojun Yin 等 11 位作者

原文摘要

LLM-based assistants on web-scale service platforms increasingly address local-service needs through multi-step planning and tool use. Under dynamic business conditions and evolving user needs, outcome-level evaluation offers limited evidence of where execution fails or how service drifts across continued interaction. We present ATLAS, a dual-horizon diagnostic evaluation framework for industrial tool-use agents. Within-turn evaluation uses trajectory-wise signals across thinking and reflection, tool and skill execution, and response generation to identify execution failures and unmet capability requirements. Across-turn evaluation uses user-wise signals spanning multiple lifecycles to diagnose failures in maintaining or updating user intent, constraints, and corrections. Fine-grained signals use rule-based checks and LLM judgments with explicit evidence scopes and decision boundaries. To support reliable diagnosis at production scale, LLM judges are calibrated against dynamically maintained references from business logs, and selected judges are distilled into compact diagnostic models. The resulting diagnostic profiles help developers attribute failures, prioritize targeted interventions, and reassess improvements. Applied in Meituan Xiaotuan, a million-scale DAU local-services assistant, ATLAS achieves high diagnostic fidelity across calibrated judges and distilled models. Blinded human auditing and a 14-day online A/B test with nearly two million users per arm confirm that the ATLAS-guided agent improves recommendation relevance, factual reliability, user engagement (+2.84% answer read-through rate), and order volume (+6.13% paid orders), leading to full production deployment.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39633HF 社区推荐

Intrinsic Distribution Matching

Deyuan Liu, Yansen Han, Kaiwen Zheng, Jun Xie, Jiacheng Liu, Zhenglin Cheng, Peng Sun, Jintao Zhang 等 10 位作者

原文摘要

One-step and few-step text-to-image generation must accurately render the details specified in the prompt despite limited opportunities for refinement. Distribution matching distillation (DMD) uses score differences to bring the generated distribution closer to a reference distribution. Following prior work, we estimate this difference with internal velocity branches, avoiding a separate fake-score network. We ask whether an explicit moment constraint improves this learned score supervision. A decomposition of the reverse-KL field into a Gaussian moment component and a non-Gaussian residual identifies statistics that can be constrained explicitly. We introduce Intrinsic Distribution Matching (IDM), which retains the full DMD correction and matches feature means and per-coordinate standard deviations using the generator's exponential-moving-average (EMA) copy, requiring no external feature encoder. Experiments use full-parameter and low-rank adaptation (LoRA) tuning on Z-Image-Turbo 6B and Qwen-Image 20B. At two network evaluations, IDM reaches 94.25% and 90.20% on Chinese and English LongText-Bench, the highest scores among the compared methods at this budget. Controlled ablations show gains from the added moment loss, while generator EMA features perform competitively with an external DINOv2 encoder across the five reported metrics. Code is available at https://github.com/inclusionAI/IDM.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2608.19491

Activation-Keyed Momentum: An Anisotropic Momentum Update via the Delta Rule

Euijin Hong, Guannan Qu

原文摘要

Most modern optimizers form their momentum as an exponential moving average (EMA) of past gradients, forgetting every direction at one fixed rate. However, the inputs a deep network sees during training can be highly anisotropic, with a few directions queried frequently while most are seen rarely. Preconditioning methods address this anisotropy by wrapping extra processing around this buffer and leave the momentum update itself unchanged. We propose Activation-Keyed Momentum (AK-Momentum), which builds direction-awareness into the momentum update rule. The gradient of a linear layer splits into an input activation that acts as a $\textit{key}$ and an output-side error that acts as a $\textit{value}$. Keying on that activation, AK-Momentum updates the momentum buffer by the canonical delta rule, so each direction is forgotten at a rate set by how often it appears. We prove that it is a valid momentum, that it applies the input-side curvature correction without matrix inversion, and that it clears stale directions faster than EMA under both a fixed and a drifting optimum. It is a drop-in replacement for the momentum buffer of any optimizer, its coefficient transfers across widths under $μ$P, and its extra compute stays between $22.2\%$ and $25.0\%$ of a gated-MLP block's linear cost with no persistent memory. In FineWeb-Edu pretraining, AdamW with AK-Momentum (AK-AdamW) reaches AdamW's validation loss in up to $46.64 \pm 4.52\%$ fewer steps at 67M and $22.14 \pm 0.81\%$ at 370M over three seeds, and the gain persists at 1B on a Chinchilla-optimal budget. A Muon baseline tuned under the same protocol sits above AK-AdamW at both language-model scales, and the gain holds for SGD, ResNet-18, and ViT-Tiny on CIFAR-10. Training-time diagnostics confirm the predicted mechanism, better gradient tracking and healthier input directions. Code is available at https://github.com/euijinh/AK-Momentum.

cs.LG · cs.CL · math.OC · stat.ML

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2607.14988

Quantum XYZ Stabilizer Codes

Alessio Baldelli, Davide Orsucci, Francisco Lázaro, Massimo Battaglioni

原文摘要

Stabilizer codes are often constructed within the Calderbank--Shor--Steane (CSS) framework, where two mutually orthogonal binary classical codes define $X$ and $Z$-type stabilizer generators. While this structure is algebraically convenient, additional non-CSS constraints may help suppress low-weight logical operators and improve decoding performance in the finite-length regime. We thus introduce quantum XYZ stabilizer codes, whose parity-check matrix (PCM) is built from three pairwise orthogonal binary PCMs associated with $X$-, $Y$-, and $Z$-type stabilizer generators. A nontrivial point is that an XYZ code instance is not automatically genuinely non-CSS: the same stabilizer group may admit a CSS generating set. We characterize this collapse, obtaining algebraic and rank conditions for deciding when the $Y$-type checks are redundant and when they define genuinely non-CSS stabilizer constraints. We also derive upper and lower bounds on the quantum minimum distance, including bounds for mixed Pauli logical operators. The novel framework includes a known non-CSS topological code, namely the XYZ$^2$ hexagonal code, and yields also sparse finite-length quantum low-density parity-check (qLDPC) constructions from intersecting-subset and quasi-dyadic code families. Simulations under depolarizing code-capacity noise and quaternary belief propagation decoding show that the proposed XYZ qLDPC instances can outperform representative CSS qLDPC instances with similar finite-length parameters.

cs.IT · quant-ph

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2512.03062

FermiGrad-SVD: An LLM Compression Pipeline with Global Rank Selection and Gauge Fixing

Roman Rausch, David Jansen, Sukhbinder Singh, Román Orús

原文摘要

Low-rank factorization of weight matrices is a promising approach to compress large language models (LLMs), but its practical effectiveness depends on allocating ranks across layers and efficiently representing the resulting factors. We introduce three complementary methods for improving SVD-based LLM compression. First, FermiGrad formulates global layer-wise rank allocation as a differentiable optimization problem by replacing discrete singular-value truncation with continuous Fermi gates. Because FermiGrad optimizes only rank parameters over precomputed low-rank factors, it avoids repeated activation decompositions and backpropagation through SVD operations. Second, the Targeted Compressor extends FermiGrad with a task-specific objective that directs limited model capacity toward a selected domain. At 60% parameter density for Llama-3.1-8B-Instruct, SciQ-targeted compression achieves 60% MMLU accuracy, exceeding the reported quantized Dobi-SVD result by 10 percentage points, while deliberately trading performance on unrelated tasks for domain specialization. Third, PivGa provides a lossless secondary compression of low-rank factors by exploiting the gauge freedom in their parametrization. The proposed framework is modular, computationally efficient, and applicable at scale, as demonstrated by compressing Llama-3.3-70B-Instruct to 50% parameter density.

cs.LG · stat.ML

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39621

The Real Cost of Context: Predicting Compute Time and Energy Consumption in LLM Prefill and Decode Phases

Lars Krupp, Daniel Geißler, Francisco M. Calatrava-Nicolas, Stefan Gerd Fritsch, Paul Lukowicz, Jakob Karolus

原文摘要

The integration of Large Language Models (LLMs) into agentic architectures has substantially increased token consumption, raising concerns about their energy footprint. Yet, estimating the energy consumption of proprietary API-based models remains challenging, and existing approaches focus primarily on generated output tokens while overlooking the impact of the large input contexts characteristic of agentic workflows. To address this gap, we systematically profile compute time and energy consumption across a wide range of prefill (input) and decode (output) token lengths. Using generalized linear models, we show that decode compute time and energy scale linearly with output length, whereas prefill compute time and energy exhibit both linear and quadratic scaling with input length. These findings demonstrate that input context becomes a relevant factor for large-context workloads. Furthermore, we show that compute time serves as a reliable proxy for energy consumption when direct energy measurements are unavailable. Our mapping framework enables precise energy estimation and supports more sustainable deployment decisions for agentic AI.

cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.26442

Exact Compilation and Resource Laws for Hidden Integer Feedback

Ganxuan Yang, Siqi Wen, Shuyan Yu, Jidong Yuan

原文摘要

Hidden rounding carries dependencies through integer memory. We study finite-memory feedback with predictable inputs that differ by one on at most $C$ releases and with every output public. A shared preparation constrains future branches until a sequence of common-input releases synchronizes the conditional registers. Preserving those constraints yields exact finite-law compilation, polynomial in support width for fixed memory and participation budgets, with constructive discrete-Gaussian transfer and calibration. A commitment hierarchy quantifies the loss from releasing dependencies early and becomes exact at depth $mC$ for memory $m$. Strict examples include three levels with one participation and a discrete-Gaussian calibration gap exceeding 0.23. Charging actual preparation time gives a privacy resource law with a square-root remainder and a leading coefficient computable in polynomial bit cost at fixed parameters. Lean refinements cover the first-order privacy analysis and binary value algorithm; separate lemmas check finite-memory algebra and paths. Certificates establish multi-memory calibration and resource effects; workload comparisons include a finite-memory prefix-error lower bound.

cs.CR

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2606.08339

Floating-point autotuning with customized precisions

Xinye Chen, Fabienne Jézéquel, Thibault Hilaire

原文摘要

Reduced-precision arithmetic offers significant opportunities to improve performance, memory usage, and energy efficiency in numerical applications, provided that numerical accuracy is preserved. This work investigates automated precision tuning through customized floating-point formats with user-defined exponent and significand sizes, enabling the emulation of emerging low-precision formats and the exploration of non-standard precision configurations within a unified mixed-precision framework. The proposed methodology, implemented in the PROMISE precision autotuning tool, combines numerical validation with a systematic search to generate program variants that satisfy user-defined accuracy requirements. To address the computational cost of this exploration, a containerized benchmarking framework supports parallel execution across multiple algorithms and parameter configurations. The approach is evaluated on a suite of numerical programs, including linear solvers and applications from the Rodinia benchmark. Results show that a substantial proportion of variables can be safely reduced to lower precision while preserving accuracy, indicating that standard double precision is often over-provisioned. These findings highlight the potential of automated precision tuning to derive efficient mixed-precision configurations tailored to application-specific accuracy requirements.

cs.MS · math.NA

查看论文
arXiv ↗PDF ↗
版本更新 · v7arXiv:2602.00667

zkCraft: Prompt-Guided LLM as a Zero-Shot Mutation Pattern Oracle for TCCT-Powered ZK Fuzzing

Rong Fu, Jia Yee Tan, Ziyu Kong, Shuning Zhang, Xianda Li, Kun Liu, Simon Fong

原文摘要

Zero-knowledge circuits enable privacy-preserving and scalable systems but are difficult to implement correctly due to the tight coupling between witness computation and circuit constraints. We present zkCraft, a practical framework that combines deterministic, R1CS-aware localization with proof-bearing search to detect semantic inconsistencies. zkCraft encodes candidate constraint edits into a single Row-Vortex polynomial and replaces repeated solver queries with a Violation IOP that certifies the existence of edits together with a succinct proof. Deterministic LLM-driven mutation templates bias exploration toward edge cases while preserving auditable algebraic verification. Evaluation on real Circom code shows that proof-bearing localization detects diverse under- and over-constrained faults with low false positives and reduces costly solver interaction. Our approach bridges formal verification and automated debugging, offering a scalable path for robust ZK circuit development.

cs.CR · cs.SE

查看论文
arXiv ↗PDF ↗