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

+0新增论文
+0其中 AI 相关
+0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:2026-11-14 15:09各信源进度 →

统计口径与数据新鲜度

统计 2026-11-15 08:09 — 2026-11-16 08:09(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:2026-11-14 15:09库内论文最新更新时间:2026-11-13 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。

23,348 已收录论文19,580 AI 相关6,729 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 23,348 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:2611.15204

Alignment De-Identifies Without Aligning: A Congruence-Invariant Relational Gap in Cross-Subject Neural Representations

Anna Notaro, Letizia Rossato, Sadasivan Puthusserypady

原文摘要

In brain-computer interfaces, inter-subject variability remains a major challenge that limits decoding performance across users. Subject-wise alignment is widely used to improve cross-subject transfer of neural decoders, yet it is typically evaluated only through decoding accuracy, which cannot determine whether aligned representations share greater task-related structure. We therefore distinguish two questions that accuracy conflates: whether alignment removes subject-specific information and whether it makes the relational organization of task conditions more similar across individuals. Using three motor imagery-EEG cohorts, trials are represented as covariance matrices and aligned by recentering each subject's mean covariance to a common reference. Subject-specific information is quantified with a subject-identification probe, whereas relational structure is assessed using representational dissimilarity matrices (RDMs) computed from condition means with the affine-invariant Riemannian distance. We show that recentering leaves this distance, and therefore the resulting RDMs, unchanged. Consistent with this result, recentering reduces subject identification accuracy from 0.96 to chance level across all cohorts, while altering RDMs only at numerical precision. Cross-subject agreement in relational structure is significantly above zero but below the estimator's lower noise ceiling. Introducing an explicit similarity regularizer during training also fails to increase held-out agreement: the apparent reduction in between-subject differences is explained by a decline in the reliability ceiling rather than improved alignment. These findings show that eliminating subject identity is not equivalent to aligning relational geometry. More broadly, alignment claims must be interpreted relative to the invariances of both the distance and the statistic used to evaluate them.

cs.HC · q-bio.NC

版本更新 · v2arXiv:2611.09364

SafeOrchestrate: Runtime Checked Intent-Driven Orchestration for Cyber-Physical Systems

Muhammad Ashfaq, Ahmed R. Sadik, Ewan Warburton, Pyry Kotilainen, Niko Mäkitalo, Tommi Mikkonen

原文摘要

Modern cyber physical systems (CPS) integrate heterogeneous sensors, actuators, controllers, and services, composed at runtime rather than at design time. Orchestrating such systems through natural language (NL) is attractive, but unconstrained large language model (LLM) tool use is risky because actions must respect device capabilities, runtime state, and safety constraints. We present Safe Orchestrate, a reference architecture for runtime checked, intent driven CPS orchestration. Given an NL intent, it derives atomic goals, generates a runtime checkable specification for each, grounds the specifications to device affordances, and verifies every candidate action against runtime state and safety constraints before execution. In a Web of Things (WoT) instantiation, the architecture produces Linear Temporal Logic (LTL) style specifications from the NL intent, grounds them to Thing Description (TD) affordances exposed through the Model Context Protocol (MCP), and checks candidate actions against satisfiability modulo theories (SMT) backed guardrails. We evaluate this instantiation on simulated smart home, greenhouse, and industrial automation tasks against a direct LLM to MCP baseline. Safe Orchestrate maintains 100% required action accuracy on every task and 100% MCP Execution Similarity (MCP ES) on all but one, where MCP ES is 92%. By contrast, the baseline degrades on multi device, event triggered, and ordered tasks, with MCP ES falling to 13%. These results suggest that combining LLM based specification generation with role aware grounding and runtime constraint checking improves reliability for open CPS orchestration without enriched TD annotations or planning domain expertise.

cs.AI

版本更新 · v2arXiv:2604.21573

CHRep: Cross-modal Histology Representation and Post-hoc Calibration for Spatial Gene Expression Prediction

Changfan Wang, Xinran Wang, Donghai Liu, Fei Su, Lulu Sun, Zhicheng Zhao, Zhu Meng

原文摘要

Spatial transcriptomics (ST) enables spatially resolved gene profiling but remains expensive and low-throughput, limiting large-cohort studies and routine clinical use. Predicting spatial gene expression from routine hematoxylin and eosin (H&E) slides is a promising alternative, yet under realistic leave-one-slide-out evaluation, existing models often suffer from slide-level appearance shifts and regression-driven over-smoothing that suppress biologically meaningful variation. CHRep is a two-phase framework for robust histology-to-expression prediction. In the training phase, CHRep learns a structure-aware representation by jointly optimizing correlation-aware regression, symmetric image-expression alignment, and coordinate-induced spatial topology regularization. In the inference phase, cross-slide robustness is improved without backbone fine-tuning through a lightweight calibration module trained on the training slides, which combines a non-parametric estimate from a training gallery with a magnitude-regularized correction module. Unlike prior embedding-alignment or retrieval-based transfer methods that rely on a single prediction route, CHRep couples topology-preserving representation learning with post-hoc calibration, enabling stable neighborhood retrieval and controlled bias correction under slide-level shifts. Across the three cohorts, CHRep consistently improves gene-wise correlation under leave-one-slide-out evaluation, with the largest gains observed on Alex+10x. Relative to HAGE, the Pearson correlation coefficient on all considered genes [PCC(ACG)] increases by 4.0% on cSCC and 9.8% on HER2+. Relative to mclSTExp, PCC(ACG) further improves by 39.5% on Alex+10x, together with 9.7% and 9.0% reductions in mean squared error (MSE) and mean absolute error (MAE), respectively.

cs.CV · q-bio.QM

新论文 · v1arXiv:2611.15201

Sparse Matrix Multiplication Exponent 1.5 in the Low-bandwidth Model

Chetan Gupta, Jukka Suomela, Hossein Vahidi

原文摘要

We study sparse matrix multiplication in the low-bandwidth model: $n$ computers multiply two $n\times n$ matrices, and each computer can send and receive one $O(\log n)$-bit message per round. The input matrices are uniformly sparse, meaning that there are at most $d$ nonzero entries in each row and column, and the task is to compute a uniformly sparse part of their product. In prior work, Gupta et al.~(SPAA 2022) presented an $O(d^{1.907})$-round algorithm over fields for this problem when the sparsity structure is globally known in advance (this is referred to as the supported setting). The upper bound was later improved to $O(d^{1.832})$ rounds. In this work, we give a deterministic algorithm that solves this problem in $O(d^{3/2})$ rounds in the supported setting, improving the previous bounds over fields and semirings.

cs.DC · cs.DS

新论文 · v1arXiv:2611.15199

Hybrid Precoding and Combining for O-RAN Massive MIMO Systems

Seraphin F. Kimaryo, Jung-Mo Moon, Kyungchun Lee

原文摘要

Recently, the radio access network (RAN) has been reimagined through open RAN (O-RAN) architecture to overcome the limitations of the traditional monolithic, black-box design philosophy and to leverage emerging technologies such as cloud computing, machine learning, and artificial intelligence. Although O-RAN disaggregates the base station (BS) into independent components to enable modular deployment, this paradigm introduces new challenges in signal processing and beamforming. In particular, beamforming in O-RAN must account for fronthaul link (FHL) capacity constraints, coordinated operation of multiple radio units (possibly from different vendors), and resource management, including transmit power allocation. To address these challenges, this work proposes novel beamforming techniques for a downlink multiuser hybrid massive multiple-input multiple-output system, in which hybrid architectures are deployed at both the users and at each BS radio unit. We formulate a sum-rate maximization problem incorporating transmit-power, unit-modulus, and FHL capacity constraints. The proposed algorithm integrates gradient descent and the Lagrange multiplier method to design analog beamformers and digital precoders, respectively, whereas digital combiners are designed to render the received streams parallel to each other. By avoiding computationally intensive operations such as eigenvalue and singular value decompositions, the proposed approach, as evidenced by simulation results, achieves superior performance with moderate computational complexity compared to existing benchmark schemes.

cs.IT

版本更新 · v4arXiv:2603.09542

NS-VLA: Towards Neuro-Symbolic Vision-Language-Action Models

Ziyue Zhu, Shangyang Wu, Shuai Zhao, Zhiqiu Zhao, Jian Zhang, Shengjie Li, Yi Wang, Anh Tuan Luu 等 11 位作者

原文摘要

Vision-Language-Action (VLA) models ground natural-language instructions in visual observations to generate actions for robotic manipulation. On long-horizon tasks, existing VLA models face structure-blind backbones, backbone-bound generalization, and flat single-objective optimization. To address these challenges, we propose a novel Neuro-Symbolic Vision-Language-Action (NS-VLA) framework. It introduces a Neuro-Symbolic Encoder that parses the instruction into a primitive plan tracked by a gated monotone stage pointer, a Neuro-Symbolic Solver that conditions a chunked VLA policy on the active sub-instruction, and Hierarchical Joint Policy Optimization, which fits the pointer on demonstration segments and refines the action policy with success-filtered online learning. Even when trained on few demonstrations, NS-VLA clearly leads on LIBERO and the held-out LIBERO-Plus (76.5% and 51.8% against 71.2% and 45.4% for its backbone), exceeds the strongest baseline by 12.7 points on long-horizon tasks, and raises real-robot success from 40.0% to 60.0%. Our code is publicly available at https://github.com/Zuzuzzy/NS-VLA.

cs.RO

新论文 · v1arXiv:2611.15197

Anytime-Valid Empirical Likelihood Inference from LLM Annotations with Adaptive Human Auditing

Hongfu Gao, Jiang Hu, Pengfei Liu, Wang Zhou

原文摘要

Large language models (LLMs) are increasingly used to annotate unlabeled data as a fast, low-cost alternative to human annotation. However, their outputs can be biased and miscalibrated relative to human labels, potentially leading to misleading analysis and inference. We propose AAELI (Adaptive Human Auditing and Anytime-Valid Empirical Likelihood Inference), an online framework for estimating the population mean of human labels under a limited auditing budget. AAELI uses LLM outputs to guide adaptive human auditing and combines them with audited labels through predictable empirical-likelihood $e$-processes to construct anytime-valid confidence sequences. Our main theoretical result shows that a predictably learned efficient projection preserves the first-order profile empirical-likelihood statistic under adaptive auditing, without requiring a prescribed convergence rate for the projection estimator. This yields an exactly anytime-valid betting construction whose continuous-mixture form incurs a scalar rather than vector-dimensional mixing penalty. Supporting results establish adaptive empirical-likelihood asymptotics and show that a learned audit rule attains the oracle variance within the stated class of augmented estimating-equation designs under the same audit-probability floor and long-run expected budget. Experiments on synthetic data and three real-world datasets demonstrate that AAELI maintains empirical anytime coverage while requiring fewer human audits than the strongest competing baseline to reach a target precision.

cs.LG

新论文 · v1arXiv:2611.15195

A General Concept-based Decomposition for Vision-Language Embeddings

Simone Alberto Peirone, Ortal Senouf, Francesca Pistilli, Giuseppe Averta, Pascal Frossard

原文摘要

Vision-Language Models are designed to capture the compositional structure of the world through multi-modal alignment; however, it remains unclear whether their representations truly align with the way a human would naturally describe the same visual input. Prior works attempt to decompose visual embeddings into their concept-level contributions, but their instance-level optimization is inherently local and does not capture the model's global semantic structure, which is essential for compositional generalization across different inputs, tasks, and domains. We introduce GRACE, a method that learns this concept-based structure at the feature space level, producing semantic and general concept decompositions. To achieve this, we propose a novel training objective that preserves the geometric relations of different samples in the concept space, enabling a better understanding of how concepts compose and relate to each other. We validate GRACE on several image and video benchmarks, showing that its concept decomposition is aligned with human perception and achieves faithful decompositions across different tasks and domains. With minimal overhead, GRACE makes existing Vision-Language Models more interpretable, while preserving most of their original performance.

cs.CV

新论文 · v1arXiv:2611.15194

The Neighbour's Label: Opposite Subgroup Effects of Same-Dialogue Exposure in Next-Act Evaluation

Ali Amini, Masoumeh Chapariniya, Rytis Maskeliunas, Robertas Damaševičius

原文摘要

Window-based next-act evaluation often aims to measure generalisation across dialogues, yet overlapping windows and similarity retrieval can expose labelled examples from the test dialogue. The nearest such example is the target's labelled predecessor, whose act is correct when the target continues it and wrong when the target changes act; a model that copies it can gain and lose in two subgroups that one aggregate averages. We induce this exposure under controlled designs on SGD, STAR and AMI, whose official splits are clean. Randomly including the predecessor's row in training raises a lexical classifier's continuation scores and lowers its transition scores in all three corpora, and purging the adjacent rows removes most of the gain. With TF-IDF-retrieved demonstrations, two 7B instruction-tuned models show the same opposite pattern in all eight model-corpus-sample arms (some single-cell intervals include zero; one replication is partial), while natural-weight accuracy changes by only -0.05 to +0.04. Exposure frequency depends on the retrieval pipeline. We evaluate two diagnostic comparisons without model retraining and find that excluding the test dialogue does not detectably reduce accuracy on SGD. All exposure is induced; no published system was re-evaluated.

cs.CV · cs.CL

版本更新 · v3arXiv:2609.35885

Local Agreement Does Not Imply Global Consensus in LLM Panels

Machiko Hirota, Akshara Nadayanur Sathis Kanna, Ujwal Kumar, Phan Xuan Tan

原文摘要

Local coordination in an LLM panel need not produce a common answer. We study 50-agent panels that update clock-time estimates after observing their neighbours. This circular response space separates agreement across the whole panel from agreement within neighbourhoods. Increasing the reasoning effort of gpt-5-mini reduces variation in local agreement, but all 20 higher-effort trials retain a one-twist configuration: neighbouring estimates remain close while the panel spans the clock. Degree-preserving rewiring instead produces global synchronisation in all 25 primary clock controls. These classifications are unchanged across 2,187 combinations of diagnostic thresholds and reporting windows. A radius-, effort-, and passage-matched scoring task also shows more uniform local agreement after rewiring, without establishing greater scoring accuracy. A simple averaging model helps interpret the clock results but predicts their configurations poorly. The experiments support measuring global agreement and local structure separately, rather than inferring consensus from uniform local agreement.

cs.MA

版本更新 · v2arXiv:2611.00176

Coinduction in Grounded Arithmetic

Bryan Ford

原文摘要

A grounded arithmetic counts a sentence true only when a terminating process establishes it; a sentence no such process settles has no value. Such a logic can state that an event happens, but not that it never happens under unbounded choice. This paper extends such a logic so that it can. The base is Reflective Grounded Arithmetic (RGA) with Inductive Grounded Arithmetic (IGA)'s least-fixed-point reading and ground induction. It adds a separate greatest-fixed-point former $ν$ with Park's coinduction rule, proves that one former cannot carry both ground induction and coinduction, and adds a dichotomy rule: the classical disjunction between a least-fixed-point atom and the greatest-fixed-point atom of its dual, under a derivable premise that the body's leaves are decided. The systems are sound for an explicit model reading every $ν$-body at its greatest fixed point, hence consistent; they agree with the least model where it has a value and prove sentences it leaves without one. The dichotomy trades the underivable hypothesis that an order is well-founded for the derivable one that its steps are decided, at the price of a change of model. With ground induction the dichotomy derives the omniscience rule of the sibling system OGA, so the system has no semantics whose certificates for universals are its own proofs; without the dichotomy it has one, though unsound for the least model. The systems prove, as single sentences, that every decided tree is well-founded or has an infinite branch given by a term, total in the new model; that random assignment terminates while its divergent twin runs forever; that a fair loop terminates and the alternating-bit protocol delivers every message; that open games are determined; that the termination gap in Nakata and Uustalu's trace semantics closes; and that Cinderella wins the bucket game at capacity three. Isabelle/HOL checks every result.

math.LO · cs.LO · cs.PL

新论文 · v1arXiv:2611.15189

WordInferBench: Can Agents Infer How to Edit from Word Document Context?

Jiayu Ding, Yupan Huang, Jinyoung Park, Yilin Jia, Wenshan Wu, Tengchao Lv, Shaohan Huang, Yuzhong Zhao 等 17 位作者

原文摘要

Document editing is among the most common knowledge-work tasks, yet existing benchmarks for evaluating autonomous agents on Word document editing provide fully specified attribute-modification instructions. They do not evaluate agents' ability to infer the appropriate attributes from document context, a capability essential in real-world editing where users expect consistency without spelling out every attribute. We introduce WordInferBench, a benchmark comprising 280 base document-editing instances, each paired with three instruction conditions: descriptive (the instruction explicitly lists every attribute change), referential (the instruction points to another object as an exemplar), and inferential (the agent must examine document context to infer what attributes to apply). To evaluate contextual attribute inference, each base instance is constructed so that the target formatting is recoverable from an in-document exemplar. The instances are derived from real Word documents collected from Common Crawl and cover 13 object types and 81 attributes. We evaluate six recent coding-agent systems and find that the best-performing evaluated system, Claude Code with Opus 4.6, achieves a 62.9% task-level pass rate on descriptive variants but only 40.7% on inferential variants, indicating that current systems complete tasks less reliably when the required formatting changes must be determined from document context.

cs.AI

版本更新 · v2arXiv:2611.08827

ActPrune: Action-Grounded Visual Token Pruning for Efficient Vision-Language-Action Models

Sungwon Moon, Minji Kim, Hyemin Boo, Daeun Song, Jiyoung Lee

原文摘要

Vision-language-action (VLA) models perform versatile manipulation tasks from language instructions, but their inference cost limits how fast a robot can respond in closed-loop control. Visual token pruning reduces this cost; however, prior methods estimate token redundancy from encoded visual embeddings or attention, only after the visual encoder has processed the entire observation. We propose ActPrune, a training-free token pruning method that moves the pruning decision ahead of the vision encoder by grounding it in a geometric prior derived from the robot's own action. The action chunk predicted at the previous control step is already available before the current observation is encoded, and it specifies the end-effector and camera motion between consecutive observations. The same action affects the two standard camera views asymmetrically: in the fixed third-person view, ActPrune retains patches along the projected end-effector trajectory together with semantic anchors, while in the wrist view, it prunes patches that remain unchanged after compensating for the action-induced camera motion. Extensive experiments on two simulation environments (LIBERO and VLABench) and four real-robot tasks demonstrate that ActPrune achieves up to 1.61$\times$ speedup and 59.5% FLOPs reduction with negligible degradation in success rates. The project site is http://mmai.ewha.ac.kr/actprune/.

cs.RO

新论文 · v1arXiv:2611.15187

OT-BLS: Optimal-Transport-based Backbone Layer Selection for Frozen DINOv3 in Cross-Domain Few-Shot Segmentation

Natsuo Okamoto, Kazuki Kobayashi, Yoshio Kato, Shuhei Tarashima

原文摘要

Cross-domain few-shot segmentation (CD-FSS) aims to segment a target class in images from a previously unseen domain using only a few annotated support examples. Training-free CD-FSS methods, which typically leverage vision foundation models such as DINOv3, achieve competitive performance without parameter updates. However, the optimal DINOv3 layer for mask prediction varies across support--query episodes. Although several layer-selection criteria have been proposed, they do not explicitly evaluate whether the predicted query foreground and background labels agree with the corresponding support labels. To address this limitation, we propose a novel layer-selection criterion, Optimal-Transport-Based Backbone Layer Selection (OT-BLS). Our key insight is that an effective layer-selection criterion can be derived from the optimal-transport correspondence between query and support patch-token features. Based on this insight, we first estimate the transport plan using a well-established optimal transport algorithm and then compute a layer-selection score by aggregating the transported mass according to the foreground--background partition defined by each candidate query mask. The resulting score reflects how well the predicted foreground and background of the query correspond to those of the support. Experiments on eight benchmark datasets demonstrate the effectiveness of OT-BLS. It outperforms the conventional last-layer baseline in 12 of 16 dataset/shot settings. Furthermore, OT-BLS improves the average mIoU from 55.72 to 57.62 and achieves a 2.99-point improvement over the best-performing existing layer-selection criterion while running 3.7--14.6$\times$ faster.

cs.CV

新论文 · v1arXiv:2611.15185

Beyond Symmetry: Asymmetric Matching with Transformers for Collective Entity Resolution

Andrei Olar

原文摘要

A core assumption of entity resolution (ER) is its symmetric nature, which rests on a principle so self-evident it borders on tautology: if a refers to the same object as b, then b must also refer to the same object as a. However, this global property of symmetry does not necessarily extend to each of the complex decision processes within ER systems. In cases involving vague or ambiguous information, enforcing symmetry can be detrimental to performance, an issue exemplified by the ``bad triplet problem''. This paper investigates the effects of input-order asymmetry within a collective ER framework that integrates distinct matching and clustering stages. The input order-aware matching stage utilizes pre-trained transformers to explicitly model and exploit asymmetry, making it possible to evaluate its effect on clustering performance across multiple benchmark datasets. Our results demonstrate that asymmetry is not only measurable but also introduces a significant skew in the final entity clusters, highlighting the need to re-evaluate supervised training methodologies for matchers, particularly the development of direction-aware training paradigms that can preserve, rather than neutralize, the inherent asymmetry in matching tasks.

cs.DB

版本更新 · v2arXiv:2611.11098

Efficient VLA Inference via Retrieval-Based Speculation and Hardware Acceleration

Zhengyang Cheng, He Zhu, Zeyu Guo, Zihan Liu, Anbang Wu, Yiming Gan, Jingwen Leng, Yu Feng 等 9 位作者

原文摘要

Vision-language-action (VLA) models enable robots to understand and execute complex instructions, but their autoregressive (AR) action generation introduces significant latency on resource-constrained robotic platforms. Speculative decoding can accelerate AR-based inference by parallelizing action generation and verification; however, existing approaches rely on costly draft models, which are inefficient on edge platforms. In this paper, we propose Zora, a hardware-software co-design framework for accelerating AR-based VLA inference. Specifically, Zora introduces a lightweight retrieval-augmented generation (RAG)-based draft model that exploits robotic dynamics without additional neural draft models. We further reduce batch verification overhead through a kinematics-guided action-acceptance strategy and a KV-cache-sharing attention mechanism. To address the retrieval overhead introduced by RAG-based drafting, we co-design dedicated hardware support, Zora core, to address the memory irregularity in RAG-based drafting. We evaluate Zora on four VLA models with the LIBERO benchmark. Zora retains comparable accuracy while achieving up to 3.4$\times$ speedup with software-only optimization and up to 7.6$\times$ speedup with hardware support over GPUs.

cs.AR · cs.RO

版本更新 · v2arXiv:2610.41405

Conditional Neural SDEs for Financial Time-Series Generation

Sina Molavipour, Alireza M. Javid, Björn Löfdahl, Mikhail Nechaev

原文摘要

Conditional generation of time-series paths is a central problem in probabilistic modeling, where the goal is to simulate future trajectories that are both realistic and responsive to external context. It has direct applications in finance and economics. Neural stochastic differential equations provide a natural framework for this task by combining continuous-time stochastic dynamics with learnable drift and diffusion functions. However, directly modeling the target state can be unstable in conditional settings, especially when the target and conditioning variables have strong level dependence, limited training data, or different dimensions. We propose a residual-anchor Neural SDE architecture that decomposes the target into a learned conditional anchor and a stochastic residual. The anchor captures the context-dependent baseline, while the Neural SDE (Stochastic Differential Equation) models the residual dynamics around this baseline. This separation reduces the burden on the stochastic dynamics and generalizes spread-based representations to more general conditioning settings. We evaluate the approach on real-world financial time series, including conditional interest-rate scenarios and a comparison with a time-series foundation model. The results show that the proposed model can generate plausible conditional paths, maintain competitive point accuracy, and provide calibrated uncertainty estimates.

stat.ML · cs.LG

新论文 · v1arXiv:2611.15184

GingivitisSeg: A Public Dataset and Comparative Benchmark for Gingivitis Segmentation in Intraoral Images

Zeynep Ozdemir, Zehra Hasgul, Irem Ulku, Omer Ozgur Tanriover, Fatma Karacaoglu, Murat Akkaya

原文摘要

Gingivitis segmentation from intraoral photographs can support computer-assisted periodontal assessment, but pixel-level delineation remains challenging due to severe foreground-background class imbalance, low-contrast boundaries, and variations in illumination and anatomy. We introduce GingivitisSeg, a public benchmark comprising 961 intraoral images with pixel-level annotations of gingival inflammation and predefined training (672), validation (144), and test (145) partitions. Seven semantic segmentation model-backbone configurations were compared using standardized partitions, input resolution, and evaluation metrics. Controlled ablation experiments examined class-weighted cross-entropy loss, Dice loss, and auxiliary supervision. UPerNet with ResNet50 achieved the highest mean Intersection over Union (mIoU) of 76.49% and gingivitis IoU of 54.91%. However, differences from DeepLabV3+ with ResNet50 and SegFormer with MiT-B2 were not statistically significant after Holm correction. The evaluated optimization components improved performance when combined in the UPerNet-ResNet50 configuration. Further analyses showed that performance was associated with gingivitis region size and decreased under stronger image blur. The dataset, benchmark splits, and source code are available at https://github.com/zeynepozdemir/GingivitisSeg.

cs.CV · cs.AI

版本更新 · v2arXiv:2611.07579

Per-Query Recall Guarantees for Approximate Nearest Neighbor Search in Dynamic Vector Databases

Woong-Kee Loh, Suan Lee

原文摘要

Vector search now runs as an operator inside database systems, where a missed neighbor makes a query answer wrong, yet approximate nearest neighbor (ANN) indexes are tuned only for a target average recall. Tuned to an average recall of 0.99, indexes left 5-19% of in-distribution queries and, without a query log, up to 99.6% of out-of-distribution queries below it (more under updates) and changed a top-5-tag aggregate over the results for up to 27% of queries. In the spirit of approximate query processing, we present PQR-Search, a $k$-ANN operator that guarantees $\Pr[\mathrm{Recall}(q) < τ] \le δ$ for every query and, for $τ< 1$, $\mathbb E[\mathrm{Recall}(q)] \ge τ$. The probability is over a random rotation inside the index alone, so no distributional assumption is needed, only that queries and updates do not depend on it, and the guarantee holds after insertions and deletions without retuning. PQR-Search prunes 1-bit codes of rotated IVF residuals with exact error quantiles and predicts its cost without data-fitted constants. On 20 datasets of up to 100M vectors, failure rates stayed at most $δ$ in every audit, including the NeurIPS'23 streaming runbook, and the aggregate changed in at most 0.04% of trials. The guarantee costs 2.5$\times$ the single-query latency of HNSW tuned as above and 1.3$\times$ lower batch throughput (geometric means). PQR-Search was 26$\times$ faster than FAISS exhaustive search in batch, faster than IVF with $n_{\rm probe}$ chosen afterwards to meet the target on every test query, and 27-76$\times$ faster than PUFFINN, which also guarantees per-query recall.

cs.DB

版本更新 · v2arXiv:2611.13597

Compute the Interactions, Not the Attributes: What a Fact Layer Must Supply in the Chinese Sexagenary System

Huixue Han, Jiaying Liu

原文摘要

Language-model systems in rule-governed domains often pair the model with a deterministic fact layer, but what that layer should compute is rarely measured. We measure it in the Chinese sexagenary system underlying BaZi chart analysis, where every fact has one table-checkable value: 682 items over twelve fact categories and 160 transit--natal collision items, each posed on a ladder of rungs that differ only in how much of the computation is done for the model. Answering directly, none of eleven open-weight models gets more than 4\% of sexagenary days or 1\% of collisions right, and supplying the attributes of every pillar does not change the collisions. A complete list of computed relations is not enough either: it leaves Qwen3-32B at 1\%, with relations listed for other pairs misattributed to the queried pair, whereas a query-filtered layer from follow-up runs, stating only the queried relations pillar by pillar, gives 92\%. On chess positions, two predictions locked in advance hold: attributes do not help, and unqueried relations are again misattributed. Three proprietary reasoning models compute 68--95\% of collisions unaided, while reasoning in smaller open models overrides supplied facts. A fact layer should compute the interactions and return the ones the question asks about.

cs.CL

版本更新 · v3arXiv:2512.05551

Automated Code Review Assignments: An Alternative Perspective of Code Ownership on GitHub

Jai Lal Lulla, Raula Gaikovina Kula, Christoph Treude

原文摘要

Code ownership is central to ensuring accountability and maintaining quality in large-scale software development. Yet, as software supply chain attacks threaten project health and quality assurance, mechanisms for assigning and enforcing responsibility have become critical. In 2017, GitHub introduced the CODEOWNERS feature, a file in the repository that names the developers responsible for specific files and automatically requests them as reviewers, to strengthen accountability and protect critical parts of the codebase. Little is known about how CODEOWNERS is adopted and practiced. We present the first large-scale empirical study of CODEOWNERS usage across over 844,000 pull requests with 1.9 million comments and over 2 million reviews. We identify 10,287 designated owners, the developers named in CODEOWNERS files, track their code review activities, and compare them with contribution-based owners, the developers who contributed at least 5% of a file's lines. Results indicate that designated owners take part in a moderate share of the pull requests that touch their files, that their collaborative behavior resembles that of contribution-based owners, and that after adoption merge time and review effort grow more slowly, or fall, relative to their pre-adoption trend. Using regression discontinuity design analysis, we further find that designated owners' review, decision (approval and merge), and authoring activity jumps upward at adoption, and that over the following year they do more review and decision work per pull request than before, whereas contribution-based owners do less review work per pull request. Our results position CODEOWNERS as a promising yet sparsely adopted mechanism for improving software governance and resilience. We discuss how projects can leverage this alternative ownership method to enhance security and accountability in open-source development.

cs.SE

版本更新 · v3arXiv:2604.24237

Computational Complexity of the Interval Ordering Problem

Simeon Pawlowski, Vincent Froese

原文摘要

We study an interval ordering problem introduced by Dürr et al. [Discrete Appl. Math. 2012] which is motivated by applications in bioinformatics. The task is to order a given set of n intervals with the goal of minimizing a certain objective which is defined via a given cost function $f$ which assigns a cost to the exposed part of each interval (that is, the pieces not covered by previous intervals). We develop a dynamic programming approach which solves the problem with $O(2^n\text{poly}(n))$ oracle calls to $f$ and arithmetic operations. Moreover, our approach yields polynomial-time algorithms for all cost functions $f$ such that $f-f(0)$ is subadditive or superadditive. This answers an open question for the function $f(x)=2^x$. We contrast these results by proving a running time lower bound of $2^{n-1}$ for any algorithm that solves the problem for every function $f$ (with oracle access only) and further proving NP-hardness for some classes of simple functions. Thus, we significantly narrow the gap regarding the computational complexity of the problem.

cs.DS

新论文 · v1arXiv:2611.15178

HyperSkill: Training-Free Omnimodal GRPO via Hypergraph-Indexed Skill-Library Evolution

Haoran Luo, Shangkai Lin, Jinyang Wu, Xinliang Zhou, Luu Anh Tuan

原文摘要

Reinforcement learning sharpens LLM reasoning, but parameter updates are expensive, forgetful, and opaque; training-free libraries swap gradients for a textual store but still use trajectory-level, correlation-based admission and flat structure. To address both, we propose HyperSkill, a Training-Free Omnimodal GRPO framework that keeps the backbone frozen and replaces gradient updates with auditable edits to a Hypergraph-Indexed Skill-Library, where skills are advantage-bearing nodes and compositions are hyperedges. It has three components: training-time distillation admits counterfactually checked skills and learns a protocol skill from failures; gated hypergraph retrieval injects advantage-ranked skills where a per-dataset gate admits them; and an execution protocol makes the frozen backbone commit to an answer. On 11 text, vision-language, and omnimodal benchmarks, HyperSkill is best or tied on every dataset and beats the best baseline by 13.0 average F1 points while keeping the backbone frozen. Our software and data are publicly available at https://github.com/LHRLAB/HyperSkill.

cs.AI

版本更新 · v3arXiv:2608.16140

Internal Reasoning and Consistency in Reflective Grounded Arithmetic

Bryan Ford

原文摘要

Reflective Grounded Arithmetic (RGA) is a paracomplete arithmetic allowing direct expression of any general-recursive computation. Every RGA term is a program and truth is grounded evaluation, with a proof certificate at each universal quantifier. This paper shows, in a machine-checked Isabelle/HOL development, that such a system is powerful enough to carry out its own metamathematics. A fuel-indexed decider of RGA's grounded semantics and checkers for RGA's proofs are compiled primitive-recursive RGA terms. RGA proves internally that its decider never affirms 0 = 1, and that every well-formed term it affirms has an RGA proof accepted by a transparent test. RGA proves that, for each valid RGA proof, its conclusion is true. RGA proves that each grounded-true well-formed term is provable. RGA proves that no proof in either of two decided classes of RGA proofs derives 0 = 1. RGA's truth predicate also serves as a medium for proof by reflection, of the kind that adds nothing to the system. RGA derives disquotation for quantifier-free formulas as rules in every context, at a free variable, and proves, as one sentence, that whenever its normalizer for polynomials in one variable gives two terms the same normal form, the equation between them is true at every value of the variable. Each such polynomial identity then follows from that sentence by closed evaluation and disquotation, with no reflection rule. A companion paper proves Gödel's incompleteness theorems for RGA and its extensions.

math.LO · cs.LO · cs.PL

新论文 · v1arXiv:2611.15172

Where Language-Model Vocabulary Rose on Stack Exchange

Fengming Liu

原文摘要

Stack Overflow banned answers generated by ChatGPT five days after the model's release, and many other Stack Exchange sites announced similar rules. I measure the vocabulary of the answers posted afterwards, 3.8 million answers on 170 sites from 2019 to 2024. A fixed list of 56 words that language models overuse rose about 14 percent above its pre-2023 trend, while the highest of 1,000 frequency-matched lists of ordinary words rose about 7 percent. In a screen of 15 site traits measured before the release, the strongest predictor of where the rise happened was the share of answers that received a body edit within 30 days of posting. On the least edited third of sites the list rose about 29 percent relative to matched controls, and on the most edited third it did not change. The gradient is also present among answers that were never edited. It follows the share of answers edited by their own authors, and the share edited by others adds little once that share is held fixed. Use of the model is unobserved, so these results cannot separate lower use on sites with a high self-edit share from any effect of editing on what the vocabulary records. Any vocabulary-based estimate for the network as a whole therefore pools sites where the marker vocabulary rose sharply with sites where it did not change.

cs.CL · cs.CY · cs.SI

新论文 · v1arXiv:2611.15170

An Experience Report on Teaching Students to Evaluate AI Agent Skills as Authorization Decisions

Zikai Alex Wen, Donald Chinn, Yan Bai

原文摘要

Cultivating a security mindset has long been a goal of computing education, and students now need it when deciding whether to trust an AI agent skill. Skills are add-ons that can shape an agent's behavior through natural language instructions and configuration settings, not just executable code. To prepare students to evaluate skills as authorization decisions, we designed a classroom activity and taught it in a senior undergraduate and a graduate cybersecurity-focused course. Students inspected three example skills, each pairing a real workflow task with one agent-specific authority, such as acting without asking for user consent before each action. The activity guided them to make initial authorization recommendations and then use a static security scanner to test and revise their recommendations. We observed in class that students needed guidance to extend their security review from a skill's code to what they would be authorizing the agent to do through the skill. Pre/post surveys from 11 volunteer respondents in this preliminary pilot showed greater self-reported confidence in evaluating skills, and post-activity responses placed more emphasis on the authority that skills carry. We release all activity materials for reuse.

cs.HC

版本更新 · v2arXiv:2611.12063

From Token Bills to Benchmark Weeks: Making Auto-Research for Data Systems Verifiable and Affordable

Kaiwen Zheng, Srijan Srivastava, Zheng Yan, Xinpeng Zhao, Shaohua Li, Chieh-Jan Mike Liang, Tianyu Chen, Haoxian Chen 等 9 位作者

原文摘要

AI auto-research is poised to forge bespoke data systems, from key-value stores and indexes to query engines, each engineered for a specific workload and hardware platform. Yet this promise rests on fragile ground. An agent optimizes what it can measure, not necessarily what users need. Natural-language requirements leave ample room for specification gaming, producing systems that blaze through benchmarks while silently losing updates or violating consistency. Worse, each research iteration consumes tokens, while validating a candidate under real concurrency, memory pressure, and failure conditions can stretch the entire research process into weeks. In this paper, we envision a more cost-efficient and verifiable approach to auto-research. Precise and composable specifications anchor the search, lightweight performance critics prune costly dead ends, and falsification and deductive verification make correctness an integral part of the process rather than an afterthought. By treating correctness, performance, research time, and token cost as equally important objectives, we aim to transform AI-generated data systems from plausible artifacts into trustworthy, production-grade systems whose behavior can be understood, audited, and ultimately relied upon.

cs.DB

版本更新 · v2arXiv:2605.12477

MEME: Multi-entity & Evolving Memory Evaluation

Seokwon Jung, Alexander Rubinstein, Arnas Uselis, Sangdoo Yun, Seong Joon Oh

原文摘要

LLM-based agents increasingly operate in persistent environments where they must store, update, and reason over information across many sessions. While prior benchmarks evaluate only single-entity updates, MEME defines six tasks spanning the full space defined by the multi-entity and evolving axes, including three not scored by prior work: Cascade and Absence (dependency reasoning) and Deletion (post-removal state). Evaluating six memory systems spanning three memory paradigms on 100 controlled episodes, we find that all systems collapse on dependency reasoning under the default configuration (Cascade: 3%, Absence: 1% in average accuracy) despite adequate static retrieval performance. Prompt optimization, deeper retrieval, reduced filler noise, and most stronger LLMs fail to close this gap. The largest gain comes from a file-based agent paired with Claude Opus 4.7 as its internal LLM, which partially closes the gap at roughly 70x the baseline cost. Code and data are available on the project page: https://seokwonjung-jay.github.io/meme-eval/

cs.LG · cs.CL

新论文 · v1arXiv:2611.15168

MAST: Medoid-Anchored Subspace Steering for Visual Grounding in Multimodal LLMs

Zhitao Chen, Zhenxin Cai, Jingyuan Li, Jin Sha

原文摘要

Existing activation steering methods for multimodal large language models (MLLMs) use the global activation mean as the steering vector, which tends to cause cross-task interference and unstable performance. We trace this failure to the multi-centered geometry of task activations. The global mean lands in the empty region between clusters and is faithful to no single task. Moreover, the steering signal in each task-pure local cluster is low-rank, and different query tasks benefit from emphasizing different families of local-basis directions under the shared routing protocol. To address this, we propose MAST (Medoid-Anchored Subspace Steering), a training-free, input-adaptive activation steering framework. MAST groups samples by task, performs multi-center clustering within each group, and selects one representative medoid per cluster. At inference time, each query is routed to multiple representative medoids. For each matched cluster, a low-rank local steering direction is extracted and aggregated across clusters to produce the final steering vector. Building on MAST, MAST-PR (Probe-Reweighted) uses a lightweight task-aware probe to reweight task-specialized local-direction families, yielding a further average improvement of +8.09 points over MAST.

cs.CV · cs.LG

版本更新 · v2arXiv:2609.34480

When Does an Image Determine the Answer? Benchmarking Visual Answerability across Charts and Scenes

Sungguk Cha, Mintae Kim, Youngsub Han, Byoung-Ki Jeon, Sangyeob Lee

原文摘要

Reliable visual question answering requires correct answers when evidence is sufficient and abstention when it is not. We introduce a benchmark of 3,000 question groups: 1,000 each from PlotQA charts, CLEVR rendered scenes, and GQA photographs. Each group keeps the question fixed across original, answer-preserving, and evidence-removing images; charts and CLEVR also include answer-changing edits. Models receive each image independently. A group succeeds only if every supported answer and every required abstention is correct. For charts and CLEVR, executable witnesses establish ambiguity under the declared task semantics: two admissible complete charts or scenes give different answers but yield the same masked observation. GQA scoring compares source labels with a permissive rule that also credits masked-image answers compatible with the full-image reference. Across 72,000 responses from six model configurations, the best complete-question success is 57.0% on charts, 43.5% on CLEVR, and 33.7% on GQA under source labels. On charts, the strongest configuration makes correct answer-or-abstain decisions on 96.2% of individual images and on all five images of 83.5% of questions. Requiring correct answers lowers success to 57.0%: in 26.5% of all chart questions, every decision is correct but at least one answer is wrong. Adding answer correctness also reverses a chart ranking based on decision accuracy. Evaluating all response obligations together reveals failures that per-image decision accuracy conceals.

cs.CV