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
+91新增 AI 导读

arXiv 最近完成扫描:2026-11-22 08:44 · 全站最近入库:2026-11-21 12:19各信源进度 →

统计口径与数据新鲜度

统计 2026-11-21 12:28 — 2026-11-22 12:28(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:2026-11-22 08:44最近发现新增版本:2026-11-21 12:19库内论文最新更新时间:2026-11-20 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。

29,158 已收录论文24,181 AI 相关8,992 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 23,345 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-22 08:42(北京时间)
新论文 · v1arXiv:2611.25908

CRAFT: Analog Circuit Design by Reasoning LLM Agents from Pre-Computed Lookup Tables

Hyunjun Park, Seungyub Han, Jason Jaesung Lee, Kahyun Kim, Soohyun Kim, Inwoo Nam, Sunmin Kim, Wooyoung Choi 等 9 位作者

原文摘要

Automated analog circuit sizing methods, from black-box optimizers to recent large language model (LLM) agents, typically search the design space with a SPICE simulator in the loop, whereas an experienced designer works out a design by hand analysis. This paper presents CRAFT, a flow in which an LLM agent designs a given topology by calculation, as a designer does, and without an initial sizing. The agent works only through DeX, a service that uses pre-computed lookup tables to provide device characteristics and to solve the whole netlist for its operating point and small-signal response. It thus designs without SPICE, and every final design is verified independently in HSPICE. We test CRAFT on four circuits in a commercial 28 nm process; the largest is a 37-transistor three-stage amplifier with 53 design variables, 20 specifications and 74 saturation constraints. With Claude Opus 5.5, CRAFT meets every specification and constraint in all 20 runs, five per circuit. On that amplifier, with DeX in place of HSPICE and an identical request for both, the same agent passes in 10 of 10 runs instead of 6 of 10 within an hour and is 2.0x faster at the median.

cs.AR

新论文 · v1arXiv:2611.25905

Validation-driven Deployment of LLM Inference Platforms for Agentic Workloads

Shai Gretz, Alon Halfon, Yoav Katz, Lena Dankin, Michal Jacovi, Rachel Tzoref-Brill, Kfir Toledo, Elron Bandel 等 13 位作者

原文摘要

Enterprises are moving agentic workloads from closed APIs to their own open-weight model deployments. For this reason, lifecycle-wide validation, from small-scale testing through sizing and pre-production to production, becomes all the more important. Typically, rollouts are evaluated with ad hoc tools whose accuracy is rarely checked. We present an end-to-end evaluation pipeline for agentic LLM serving and empirically ground the design choices that connect its stages. Our experiments show that replaying captured traces reproduces the per-session load of the original agent sessions: at the 95th percentile replay stays within 1.18x of the recorded input tokens while live re-runs reach 4.98x, so a small set of recorded sessions yields a repeatable per-session token load. A code-coverage analysis of vLLM further shows that the evaluated live, replayed, and synthetic workloads are complementary in this configuration: each exercises Python serving paths not reached in the measured runs of the others. In a case study rolling out four large mixture-of-experts (MoE) models on llm-d, experiments show that the target models reached about half the optimal concurrency, and correspondingly about 60% of the throughput, on captured workloads compared with public traces. We open-source the relevant tools and release a trace dataset.

cs.AI

版本更新 · v4arXiv:2505.10006

Improved Rank Aggregation under Fairness Constraint

Diptarka Chakraborty, Himika Das, Sanjana Dey, Alvin Hong Yao Yan

原文摘要

Aggregating multiple input rankings into a consensus ranking is essential in various fields such as social choice theory, hiring, college admissions, web search, and databases. A major challenge is that the optimal consensus ranking might be biased against individual candidates or groups, especially those from marginalized communities. This concern has led to recent studies focusing on fairness in rank aggregation. The goal is to ensure that candidates from different groups are fairly represented in the top-$k$ positions of the aggregated ranking. We study this fair rank aggregation problem by considering the Kendall tau as the underlying metric. While we know of a polynomial-time approximation scheme (PTAS) for the classical rank aggregation problem, the corresponding fair variant only possesses a quite straightforward 3-approximation algorithm due to Wei et al., SIGMOD'22, and Chakraborty et al., NeurIPS'22, which finds closest fair ranking for each input ranking and then simply outputs the best one. In this paper, we first provide a novel algorithm that achieves $(2+ε)$-approximation (for any $ε> 0$), significantly improving over the 3-approximation bound. Next, we provide a $2.881$-approximation fair rank aggregation algorithm that works irrespective of the fairness notion, given one can find a closest fair ranking, beating the 3-approximation bound. We complement our theoretical guarantee by performing extensive experiments on various real-world datasets to establish the effectiveness of our algorithm further by comparing it with the performance of state-of-the-art algorithms.

cs.DS

新论文 · v1arXiv:2611.25904

Multiple-Camera Tracking of Rigid Objects

Frédérick Martin, Radu Horaud

原文摘要

In this paper we describe a method for tracking rigid objects using one or several cameras. The tracking process consists of aligning a 3-D model representation of an object with image contours by measuring and minimizing the image error between predicted model points and image contours. The tracker behaves like a visual servo loop where the internal and external camera parameters are updated at each new image acquisition. We study in detail the Jacobian matrix associated with this minimization process in the presence of both point-to-point and point-to-contour matches. We establish the minimal number of matches that are needed as well as the singular configurations leading to a rank-defficient Jacobian matrix. We find a mathematical link between the point-to-point and point-to-contour cases. Based on this link we show that the latter has the same kind of singularities than the former. Moreover, we study multiple camera configurations which optimize the robustness of the method in the presence of single-camera singularities, bad, noisy, or missing data. Extensive experiments done with a complex ship part and with up to three cameras validate the method. In particular we show that the tracker may well be used as a camera calibration procedure.

cs.RO · cs.CV

新论文 · v1arXiv:2611.25903

From Cases to Criteria: Reusing Evaluation Knowledge for Task-Specific Rubric Generation

Qiyan Deng, Long-Kai Huang, Kangfei Zhao

原文摘要

Query-specific rubrics make LLM evaluation more fine-grained and transparent, and generating high-quality rubrics requires both identifying the target query's evaluation scope and specifying concrete checks. Historical rubrics provide reusable evaluation experiences, yet existing approaches reuse either complete query-rubric cases or aggregated criteria and dimensions without explicitly connecting recurring evaluation requirements to their criterion-level evidence and source-query context. This can preserve irrelevant task details or lose the broader evaluation structure. In this paper, we introduce RubricDuo, which organizes historical query-rubric pairs into a Rubric Knowledge Graph linking evaluation dimensions, shared concepts, local concepts, criteria, and queries, where reusable evaluation requirements across queries are abstracted by shared concepts. Given a target query, RubricDuo first combines concepts relevant to the target query with retrieved historical cases to draft a rubric with broad coverage. It then revisits each draft criterion, retrieves concept-linked historical criteria and their source queries, and adapts applicable evidence to refine the criterion. Experiments on RubricBench and HealthBench demonstrate that RubricDuo achieves the overall best performance among the evaluated automatic methods in both alignment with human-authored rubrics and downstream LLM judging.

cs.AI

新论文 · v1arXiv:2611.25902

Does It Depend on the Split? Frozen Satellite Features vs. Fine-Tuned CNNs for Tree Species Mapping

Dharani Deivasihamani, Nandakishore Puttashamachar

原文摘要

Self-supervised foundation models are often said to need fewer labels than supervised training. We test this on a forestry task: classifying 20 tree species classes in 0.2 m aerial images from the TreeSatAI dataset. We compare a frozen DINOv3 model pretrained on 493 million satellite images, with only a linear classifier trained on top, against an ImageNet ResNet-50 that is fine-tuned end to end. On the official split, fine-tuning wins clearly (macro F1 0.764 vs. 0.685). However, 28.8% of the official test patches overlap a training or validation patch, and copying the label of the nearest such patch is correct for 83.6% of them. We therefore build a spatial split that holds out whole 2 km squares with a 200 m buffer. On this split the comparison becomes inconclusive (0.648 vs. 0.640, paired 95% CI of the difference [-0.054, 0.016]), and the learning curves of the two models overlap at every label budget from 343 to 33,126 training patches. On the five rarest classes, the frozen model achieves a higher mean F1 (0.566 vs. 0.445), although the confidence interval for this difference includes zero ([-0.05, 0.23]). Diagnostics show that the frozen features mainly miss differences between closely related species, such as the two native oaks. Choosing the input size and layer on validation data (224 px, block 16) raises the frozen model to 0.662 on the spatial split. A comparison that looks decisive on the official split is thus inconclusive under geographically separated evaluation. We release the spatial split and our code.

cs.CV

新论文 · v1arXiv:2611.25898

Verifiable Logic in Medical LLMs: A Critical Review of Agentic Clinical Reasoning Optimization and Neuro-Symbolic Grounding

Youssef ElNahas, Adel Ammar, Soumaya Chaffar, Wadii Boulila

原文摘要

Clinical large language models are usually judged by whether the final answer is right. Yet a model can reach a fluent, plausible output through invalid reasoning that static benchmarks do not detect, and confident hallucination, causally invalid inference, and spurious demographic shortcuts can pass conventional evaluation. This critical review covers four paradigms in turn: how clinical hallucination arises and why static metrics miss it; the agentic hallucination cascade, in which multi-agent and tool-augmented systems propagate and amplify early reasoning errors; process reward modeling as step-level supervision; and neuro-symbolic grounding, in which formal logic engines, ontology reasoners, and knowledge graphs place verifiable constraints on neural output. Within their encoded domains, systems with symbolic veto layers offer guarantees that hold relative to their encoded knowledge, and they report large gains on the safety measures they use. These results come from single studies, mostly in simulated or small-sample settings, and are not directly comparable with neural baselines. Process reward models remain bounded by the statistics of their training data. We propose a conceptual framework that maps each agentic failure mode to a symbolic intervention point, and we argue that clinical AI safety depends substantially on system architecture. Because current regulation emphasizes performance and does not require verification of reasoning steps, we argue that responsible certification should also audit reasoning trajectories alongside accuracy on static benchmarks.

cs.CL · cs.AI · cs.CY · cs.LG

新论文 · v1arXiv:2611.25895

One Schedule Does Not Fit All: Adaptive Inference Schedules for Diffusion Language Models

Rabimba Karanjai, Linsey Pang, Hemanth Hegadehalli Madhavarao, Richard Williamson, Uma Kona, Prakhar Mehrotra

原文摘要

Diffusion language models typically use a fixed denoising schedule across workloads. We show that the cheapest schedule that preserves quality depends on the task, model, and sampler. We propose Task-Calibrated Denoising Allocation (TCDA), a deployment rule with an explicit error guarantee. Given candidate samplers, paired calibration data, a utility tolerance, and guardrails against degeneration, TCDA uses simultaneous one-sided bootstrap bounds to certify which candidates meet both the utility and guardrail criteria. It deploys the cheapest certified candidate or falls back to the full schedule. The probability of deploying an infeasible schedule is asymptotically at most alpha. Candidates include early prefix decoding, state-reusing continuation, reduced time grids, and confidence-threshold decoding. Across seven datasets and two model families, TCDA certifies reductions in denoiser evaluations by factors of 3.4 on XSum and 2.0 on WMT14 for ELF-B; 8.0 on WMT14 and 4.1 on GSM8K for LLaDA-8B; and 1.33 on OpenWebText for both ELF-M and ELF-L. A validation protocol fixed before generation confirmed every selection on fresh seeds. Across 500 calibration splits, false certification remained below 2% in every setting at a nominal level of 5%. Certified schedules vary across tasks and model scales. Trajectory measurements suggest that when decoded tokens commit is more informative for schedule selection than the remaining latent discretization error. We also find that the ELF reference sampler draws its time grid at random, so reproducible schedule comparisons for unconditional generation require fixing the grid.

cs.AI · cs.CL · cs.LG

新论文 · v1arXiv:2611.25894

Explainable AI for Solar PV Power Forecasting: From Model Selection to Trustworthy, Reproducible Deployment

Nuzhat Khan, Muhammad Paend Bakht, Ab Al-Hadi Ab Rahman, Fabiano Pallonetto, Amy Fahy, Mohamed Khalafalla Hassan, Paul Stynes, Muhammad Anique Aslam 等 10 位作者

原文摘要

Machine learning (ML) is now a standard tool for forecasting the output of solar photovoltaic (PV) plants, yet published accuracy figures are often hard to compare and harder to trust. Many studies validate with random splits, report no reference forecast, and present feature attributions without saying what they can and cannot show. This paper reviews the main families of ML methods for PV power forecasting and resource assessment, from tree ensembles to deep and foundation models, and sets out a reproducible workflow that combines automated model search with explainable AI. The workflow is then tested on an openly available dataset: a 16.32 kWp plant in Porto, Portugal, forecast one day ahead from a 4 km numerical weather prediction grid. Over a full test year, the best models reach a normalised root mean square error (nRMSE) of 11.6% of installed capacity, about half the error of smart persistence. The choice of input features changed accuracy far more than the choice among strong algorithms, and shuffled cross-validation was mildly optimistic. After conformal calibration, the nominal 80% prediction intervals covered 81.7% of observations, against 64.5% before. SHAP rankings stayed stable when the model was retrained on different periods but shifted when correlated inputs were removed, so attributions should be read as properties of a model rather than of the plant. The paper closes with practical lessons and open problems, including deployment on edge hardware and the effect of a changing climate.

cs.AI

版本更新 · v2arXiv:2609.32355

Gradients for Interventions and Activation Values for Detection: Targeted Feature Learning in Language Models

Jonathan Drechsel, Steffen Herbold

原文摘要

Model-internal features can be studied through both their ability to identify a specified concept and their causal effect when manipulated, e.g., through steering or weight editing. A prominent approach to feature learning is Sparse Autoencoders (SAEs), which learn broad feature dictionaries whose relation to particular concepts is typically identified post hoc. However, many interpretability questions are instead hypothesis-driven and concern a concept specified in advance. We study this setting as targeted feature learning, where a single feature is constructed for such a predefined concept. We present a controlled comparison across three model signals (activation values, activation gradients, and parameter gradients) and two estimators (contrastive mean and a learned one-dimensional encoder-decoder), yielding six targeted methods, with CAA and GRADIEND as existing instances and four new methods covering the remaining combinations. We compare these methods against pretrained SAEs across 15 tasks and ten language models, evaluating both detection and causal intervention. Across models, the strongest detection performance is achieved by contrastive activation value methods, whereas the strongest intervention performance is achieved by gradient signals, most strongly with a contrastive mean. Notably, within the contrastive mean family, activation gradients substantially improve intervention over activation values without requiring parameter space interventions. Overall, our results show that targeted feature quality depends jointly on the model signal and estimator, with detection and intervention capturing complementary properties.

cs.CL · cs.LG

新论文 · v1arXiv:2611.25889

EvoMem: Self-Evolving Skills for Time-Aware Long Video Memory Agents

Yutong Guan, Meixing Shi, Haoran Sun, Bing Sun, Yuxiang Cai, Yankai Jiang

原文摘要

Although memory-augmented architectures have expanded the ability of large vision-language models (LVLMs) to process long videos, existing retrieval processes still largely rely on global semantic matching and lack state-driven temporal control, often returning fragmented or redundant evidence and failing to maintain a coherent evidence chain. To address this challenge, we propose EvoMem, a time-aware long-video memory agent guided by self-evolving retrieval skills. EvoMem constructs a hierarchical time-indexed memory, where different types of evidence are aligned onto a unified video timeline; during retrieval, temporal cues from acquired evidence are further converted into explicit signals that constrain subsequent memory access. To address recurring procedural errors in long-video question answering, such as unclear retrieval targets, improper memory access paths, or premature stopping, EvoMem introduces retrieval skills to provide process-level guidance, enabling the agent to dynamically adjust retrieval content, memory operations, and target temporal ranges according to the current evidence state, while deciding when to stop. Furthermore, inspired by reinforcement learning, we formulate skill evolution as a reward-guided policy improvement process. Each candidate skill is treated as an external policy patch, and rewards are assigned according to answer-correctness state transitions on the validation set. In this way, effective skills are accepted and low-quality skills are rejected without updating model parameters. Experiments on four long-video question answering benchmarks show that EvoMem significantly outperforms existing baselines, achieving an average accuracy of 60.1% and an average improvement of 8.6 percentage points over the previous best methods, demonstrating its effectiveness for long-video reasoning.

cs.CV

版本更新 · v2arXiv:2604.00853

MotionGrounder: Grounded Multi-Object Motion Transfer via Diffusion Transformer

Samuel Teodoro, Yun Chen, Agus Gunawan, Soo Ye Kim, Jihyong Oh, Munchurl Kim

原文摘要

Motion transfer enables controllable video generation by transferring temporal dynamics from a reference video to synthesize a new video conditioned on a target caption. However, existing Diffusion Transformer (DiT)-based methods either transfer the motion of a single object or handle multiple objects without binding each object caption to its object, restricting fine-grained control in real-world scenes. In this work, we introduce MotionGrounder, a training-free DiT-based framework that is the first to jointly handle motion transfer and object grounding in multi-object scenes. Our Flow-based Motion Signal (FMS) in MotionGrounder provides a stable motion prior for target video generation, while our Object-Caption Alignment Loss (OCAL) grounds object captions to their corresponding spatial regions. We further propose two new metrics: the Object Grounding Score (OGS), which jointly evaluates (i) spatial alignment between source video objects and their generated counterparts and (ii) semantic consistency between each generated object and its target caption, and the Motion Grounding Score (MGS), which evaluates whether each object moves like its source object at the correct location. Our experiments show that MotionGrounder improves object grounding over recent baselines, with the advantage appearing once several objects are present, across quantitative, qualitative, and human evaluations.

cs.CV

版本更新 · v3arXiv:2609.39131

Characterizing High Bandwidth Flash for LLM Serving

Zack Yu, Chloe Wong, Coleman Hooper, Minjae Lee, Wonjun Kang, Youngjin Cho, Michael W. Mahoney, Yakun Sophia Shao 等 10 位作者

原文摘要

Large language model (LLM) serving requires substantial memory to store model weights and KV caches. Agentic workloads compound this pressure through repeated interactions over growing contexts, making it important to retain KV cache for reuse. High-bandwidth flash (HBF) offers a way to expand accelerator memory capacity for LLM serving, but its access costs and limited write endurance complicate its use. We evaluate HBF for high-throughput agentic serving across system design and scheduling choices to understand when additional capacity improves serving performance and energy efficiency. We introduce an HBM-HBF-host hierarchical storage system and buffered cache-aware scheduling, and use trace-driven simulations to analyze their effects on performance, energy consumption, and HBF write lifetime. Across the evaluated workloads, the fastest HBF-augmented systems reduce completion time by 21.1-86.9% relative to HBM-only systems. Modeled energy savings reach 62.4%, with benefits depending on the workload and weight placement. Buffered cache-aware scheduling extends estimated HBF write lifetime from 0.99 to 13.63 years in the evaluated configuration. These results demonstrate the importance of coordinating data placement and scheduling in serving LLM with HBF, positioning HBF as an actively utilized lower-level memory tier rather than passive storage.

cs.LG · cs.AR · cs.DC

版本更新 · v2arXiv:2609.35628

Arbitrary-Accuracy Neural Approximation with Optimal Neuron Count and Optimal Bit Complexity

Zilan Cheng, Li-Lian Wang, Zhongjian Wang

原文摘要

We study the minimum number of hidden neurons required for arbitrary-accuracy approximation of continuous functions on $[0,1]^d$ and the associated encoding complexity. For $d\geq 2$, we construct two fixed, explicitly defined activation functions and, for each, a closed-form network with two hidden layers of widths $d$ and $1$ that achieves arbitrary accuracy in the uniform norm. We prove that $d+1$ is the exact minimum total number of hidden neurons among standard feedforward networks with locally integrable activations and affine outputs. We further give a simpler construction using a single elementary activation that combines the floor and exponential functions. This construction requires three hidden layers of widths $d$, $1$, and $2$. If a skip connection is allowed, widths $d$, $1$, and $1$ suffice. For a bounded $α$-Hölder class with $α\in(0,1]$, the networks with the digit-extraction activation and with the elementary activation require $O(\varepsilon^{-d/α}\log(1/\varepsilon))$ bits, matching the metric-entropy lower bound up to a logarithmic factor. The other $(d,1)$ network, which stores differences of neighboring values along a spanning tree, needs only $Θ(\varepsilon^{-d/α})$ bits and is optimal in both neuron count and bit complexity.

cs.LG · cs.NE · stat.ML

新论文 · v1arXiv:2611.25884

Trustworthy Geospatial AI Workflows: Reproducible Maritime Analytics and Forecasting on Open-Source AIS Data

Divya Acharya, Thomas M. Roehr, Arnaud Gotlieb

原文摘要

Publicly available geospatial data provide increasing opportuni- ties for data-driven maritime analytics and predictive modelling. However, their availability does not necessarily imply that the underlying observations are equally suitable for downstream AI applications. Properties of the data source and observation pro- cess, including provenance, reporting regularity, spatial coverage, observation density, and missingness, can affect how geospatial data are interpreted, processed, and ultimately used by predictive models. At the same time, trustworthy geospatial AI requires ana- lytical workflows in which these processing decisions can be made explicit and reproduced. This paper presents a reproducible mar- itime geospatial AI workflow built around damast, an open-source data-processing and machine-learning library developed at Simula Research Laboratory. Using AIS observations from multiple public data providers as a case study, we demonstrate metadata-aware validation, reusable processing pipelines, maritime trajectory con- struction, and reproducible machine-learning experimentation. The predictive stage is model-agnostic and is instantiated in this study using a Transformer-based vessel-trajectory forecasting model. Building on these implemented capabilities, we further outline an envisioned extension toward trustworthiness-aware geospatial AI. The extension introduces explicit characterization of source- and data-level properties and their relationship with downstream model performance. This distinguishes capabilities demonstrated in the current workflow from trustworthiness-oriented components identified for future development. Overall, the study demonstrates how reproducible open-source infrastructure can provide a practi- cal foundation for developing more transparent and trustworthy geospatial AI workflows.

cs.AI

新论文 · v1arXiv:2611.25880

Evolving Decoding Algorithms for Quantum Error Correction

Eduardo Dueñez, Edgar Said Hernandez Sanchez, Edgar A. Duéñez-Guzmán, Suzanne Sadedin

原文摘要

Quantum computation entails robustness engineering: a noisy quantum state is measured continuously, and a classical algorithm (the decoder) must infer corrections that reliably remove the noise, fast enough to keep up. We write several published belief-propagation (BP) decoders for quantum low-density parity-check (qLDPC) codes in a Domain-Specific Language (DSL), and search over it with grammatical evolution. Written in the language, the published decoders show their parameters in plain view, allowing us to re-parameterise BP+OSD (ordered statistics decoding) and BP+LSD (localized statistics decoding) decoders with damping, before any search, making each 17.1--41.0$\times$ faster at a higher error rate, and BP+OSD-0 about 103 faster than the ldpc package's version tuned on the [[72,12,6]] bivariate-bicycle code. Evolution over the language finds decoders that improve the time-error front. One makes 23% fewer logical errors than BP+OSD-0 with damping at 0.64$\times$ its time, and decoders evolved from random code are more accurate than ldpc's tuned BP+OSD-0 at 1/31 of its time or less. Frozen at one operating point, the two fastest evolved decoders generalize well across noise levels, code sizes and round counts, staying at the time-error front. Ultimately, we demonstrate that capturing expert knowledge within a DSL creates a powerful intermediate layer---one that effectively bridges human intuition and machine-driven optimization to unlock highly efficient quantum error correction.

quant-ph · cs.AI · cs.MA · cs.NE

版本更新 · v2arXiv:2602.08505

Are Vision Foundation Models Foundational for Electron Microscopy Image Segmentation?

Caterina Fuster-Barceló, Virginie Uhlmann

原文摘要

Although vision foundation models (VFMs) are increasingly reused for biomedical image analysis, it remains unclear whether the latent representations they provide are general enough to support effective transfer and reuse across heterogeneous microscopy image datasets. Here, we study this question for the problem of mitochondria segmentation in electron microscopy (EM) images, using two popular public EM datasets (Lucchi++ and VNC) and three recent representative VFMs (DINOv2, DINOv3, and OpenCLIP). We evaluate two practical model adaptation regimes: a frozen-backbone setting in which only a lightweight segmentation head is trained on top of the VFM, and parameter-efficient fine-tuning (PEFT) via Low-Rank Adaptation (LoRA) in which the VFM is fine-tuned in a targeted manner to a specific dataset. Across all backbones, we observe that training on a single EM dataset yields good segmentation performance (quantified as foreground Intersection-over-Union), and that LoRA consistently improves in-domain performance. In contrast, training on multiple EM datasets leads to severe performance degradation for all models considered, with only marginal gains from PEFT. Exploration of the latent representation space through various techniques (PCA, Fréchet Dinov2 distance, and linear probes) reveals a pronounced and persistent domain mismatch between the two considered EM datasets in spite of their visual similarity, which is consistent with the observed failure of paired training. These results suggest that, while VFMs can deliver competitive results for EM segmentation within a single domain under lightweight adaptation, current PEFT strategies are insufficient to obtain a single robust model across heterogeneous EM datasets without additional domain-alignment mechanisms.

cs.CV

版本更新 · v3arXiv:2602.18140

Flex-Plorer: A Precision-Aware Hardware-Software Co-Design Framework for an Edge SNN Accelerator

Mohammad Farahani, Mohammad Rasoul Roshanshah, Saeed Safari

原文摘要

Neuromorphic accelerators promise high energy efficiency and computational density for spiking neural networks (SNNs), especially in wearable devices and neural prosthetics where power constraints are stringent. However, most existing platforms exhibit rigid architectures with limited configurability, which restricts their adaptability to heterogeneous signals and diverse design objectives. To address this gap, we present Flex-Plorer, a precision-aware hardware-software co-design framework using heuristic search over weight and decay precisions for a fixed architecture. It couples hardware-aware fixed-point simulation with synthesis-trained LUT/FF estimators and analytical BRAM estimation to guide user-defined accuracy-resource trade-offs without per-candidate synthesis. The best-found configuration generates specialized RTL, drivers, and deployment data. Its substrate, Flexi-NeurA, supports multiple neuron models and topologies through event-driven, time-multiplexed cores with shared neuron datapaths and address-event communication. Design-time component selection and automatic memory sizing specialize the implementation without manual RTL modification. Comprehensive evaluations demonstrate the versatility of the proposed flow across three application domains: (i) auditory processing, achieving 75% accuracy with an energy consumption of 1.48mJ per inference; (ii) dynamic vision sensor gesture recognition, achieving 93% accuracy at 0.55mJ; and (iii) standard vision classification, achieving 97% accuracy at 0.12mJ. Together, Flex-Plorer and Flexi-NeurA form an application-adaptive, energy-efficient co-design flow for deploying SNNs across heterogeneous edge-AI workloads.

cs.AR · cs.NE

新论文 · v1arXiv:2611.25879

Coherent belief and opinion propagation produces more echo chambers: corrected version

Hiro Kataoka, J{é}r{ô}me Euzenat, Koji Hasebe

原文摘要

The occurrence of echo chambers is a well-known phenomenon in social networks. Echo chambers are groups of agents satisfying three properties: homogeneity, i.e., sharing the same opinion, segregation, i.e., listening primarily to agents of the same group, and reinforcement, i.e., increasing opinion similarity over time. Opinion dynamics models have been used to explain how echo chambers occur through agents updating their opinions. However, social network users also exchange beliefs supporting their opinions. Here, we aim at identifying the effect of propagating beliefs together with opinions on the formation of echo chambers. For that purpose, we complement an existing opinion dynamics model with classical belief merge operations. Beliefs and opinions are also maintained coherent guided by values that agents share. We introduce measures, applying to either beliefs and opinions, to identify echo chambers based on the three properties. This model is used to simulate opinion and belief propagation with varying initial conditions and agent tolerance for different opinions/beliefs. Results confirm the occurrence of echo chambers observed in previous opinion dynamics studies in a different setting and with a stricter measure. They also demonstrate the occurrence of echo chambers with belief propagation. Finally, they show that connecting opinions to beliefs and maintaining their coherence increases the number of belief echo chambers, but that they involve a reduced number of agents.

cs.AI · cs.MA

新论文 · v1arXiv:2611.25877

On the trajectories of the ID algorithm

Ivano Lodato, Simran Hiranandani, Aditya Varna Iyer

原文摘要

The ID algorithm decides whether an interventional distribution can be computed from observational data and, when it can, returns a formula in which products, ratios and summations of conditional probabilities are nested inside one another. We study these formulas by following the trajectories of the calls of the algorithm, each of which ends with a tail (an alternation of lines 7 and 2) and an exit, and we derive two main results. First, line~4 and the exit lines fix the single outer summation of the output, while the tail forms everything inside it, both the inner summations and the ratios. Second, a ratio keeps a summation in its denominator only when a repetition of the tail is entangled, and a test of ancestry in the original graph decides, before any formula is computed, whether a topological order can be chosen so as to avoid it. We also derive a bound on the nesting depth of the output, and show a simple ``dimensional'' count that every output has to satisfy. Together, these results constrain the form of an ID output and guide its implementation and simplification.

stat.ME · cs.AI · math.PR

版本更新 · v3arXiv:2305.11741

Proving Almost-Sure Innermost Termination of Probabilistic Term Rewriting Using Dependency Pairs

Jan-Christoph Kassing, Jürgen Giesl

原文摘要

Dependency pairs are one of the most powerful techniques to analyze termination of term rewrite systems (TRSs) automatically. We adapt the dependency pair framework to the probabilistic setting in order to prove almost-sure innermost termination of probabilistic TRSs. To evaluate its power, we implemented the new framework in our tool AProVE.

cs.LO

版本更新 · v2arXiv:2609.37851

FlowMap-OPD: On-Policy Distillation for Few-Step Flow-Map Generators

Zhiqi Li, Cheng Wan, Bo Zhu

原文摘要

Few-step flow-map generators, such as MeanFlow and consistency models, sample in a few long-range jumps. On-policy distillation (OPD) trains a student on states drawn from its own sampler, but existing methods compare teacher and student through the local stochastic transitions of a multi-step sampler, which a flow-map student never deploys. We present FlowMap-OPD, the first systematic study of OPD for flow-map students. It separates state collection from supervision: the student collects states with its native few-step sampler, while the teacher supervises the flow map, velocities induced from it, or the instantaneous velocity linked to the map by a separately weighted consistency penalty. We find that instantaneous-velocity supervision with a small consistency weight, our chosen form, transfers the teacher's gains best, whereas supervising the deployed map, although the natural choice, transfers them worst when teacher and student start from different pretrained models. This gives a counter-intuitive rule: \emph{sample with the flow map, learn the velocity}. As a further benefit, this form applies to any teacher that defines a velocity, including multi-step ones, and to students initialized from multi-step models. On FLUX and SD3.5, a single few-step student consolidates three specialist teachers within 300 updates, matches or exceeds each on its own benchmark (e.g., OCR 0.883 versus 0.850 on FLUX), and outperforms both multi-reward reinforcement learning and LoRA merging.

cs.CV · cs.LG

新论文 · v1arXiv:2611.25873

WHU-Road3D: A dataset and benchmark towards fine-grained road surface mapping

Xiaoxin Mi, Zhen Cao, Chong Liu, Yuhao Li, Zhen Dong, Bisheng Yang, Yang Jia, Shengwu Xiong

原文摘要

Accurately extracting and mapping road surface elements is crucial for applications such as high-definition maps, road asset management, and intelligent transportation systems. This study first provides a comprehensive review of the benchmark datasets for extracting and mapping road elements, encompassing overhead-to-perspective data, large-to-local scales, and coarse-to-fine details. Subsequently, we thoroughly review the relevant approaches for extracting road surface elements. However, we argue that existing datasets still require enhancements in semantic richness, data modality diversity, and scenario complexity. To bridge these gaps, this paper presents WHU-Road3D, a large-scale multimodal dataset for fine-grained road surface mapping to address these shortcomings. WHU-Road3D provides approximately 100 km of diverse Chinese highways and urban streets. It features a rich combination of MLS point clouds, associated panoramic images, vehicle trajectories, detailed vectorized annotations, and pointwise instance annotations for road surface elements. The dataset carefully annotates 15 distinct categories of road elements, including common markings like road boundaries, zebra crossings, lane dividers, stop lines, as well as 10 types of direction symbols. To our knowledge, WHU-Road3D stands as the first multimodal dataset for mapping such a comprehensive range of road surface elements, designed to facilitate cutting-edge research in instance segmentation, full-element extraction and mapping on road surfaces. Beyond these immediate applications, WHU-Road3D also provides a valuable source for hierarchical road topology reasoning, map updating, and street scene generation. The dataset, along with the code for our benchmark methods, will be made publicly available at \href{https://github.com/WHU-USI3DV/WHU-Road3D.git}{https://github.com/WHU-USI3DV/WHU-Road3D.git}.

cs.CV

版本更新 · v3arXiv:2611.12805

2D Gum-Net: Unsupervised Dense Elastic Image Registration via Symmetric Correlation and Spectral Pooling

Aarya Bhave, Anushka Ajay Bhave, Akanksha Narula, Marius Sebastian Shokoui, Alina Hafner

原文摘要

Dense non-rigid registration, the recovery of a pixel-wise deformation field that warps one image into geometric agreement with another, is fundamental across medical imaging, structural biology, and correspondence estimation, yet learning-based approaches face a severe bottleneck: ground-truth deformation fields are notoriously difficult to obtain for real data. We present 2D Gum-Net, an end-to-end architecture that predicts dense elastic deformation fields fully unsupervised. We transfer the Geometric Unsupervised Matching Network (Gum-Net), originally developed for 3D subtomogram alignment in cryo-electron tomography, to 2D elastic registration: all three modules are redesigned in 2D, and the rigid transformation head is replaced by a Dense Spatial Transformer that regresses a sparse control-point grid, lifted to a full-resolution field $Φ$ by bicubic interpolation. Downsampling uses DCT spectral pooling, implemented as frozen-basis matrix multiplications ($\approx 50\times$ GPU speedup), preserving the fine, quasi-periodic structure that max and average pooling destroy. We evaluate on contact-based fingerprints, a challenging testbed for this task: locally self-similar ridge patterns make correspondence maximally ambiguous, skin pressed against a sensor generates authentic large-displacement elastic distortion, and minutiae act as precise landmarks through which an independent matcher externally verifies the topological correctness of predicted fields. Trained purely on synthetic impressions and evaluated zero-shot on real benchmarks, our method improves structural correlation by up to 15.7% on FVC2004 and raises genuine match scores under an independent matcher while leaving imposter scores essentially unchanged. This suggests that the predicted deformations restore true structure without fabricating correspondences.

cs.CV · cs.AI

新论文 · v1arXiv:2611.25870

X Hedging: towards an eXplainable artificial intelligence option Hedging framework

Bjørn André Aaslund, Johannes Berge, Ayoub Haida, Ying Ni, Rita Pimentel

原文摘要

We propose X Hedging, an alternative to the Deep Hedging framework that addresses computational and explainability limitations of neural-network-based hedging strategies. Deep Hedging leverages neural networks to learn flexible hedging policies but relies on opaque models and typically requires large training datasets, which can complicate interpretation and limit its effectiveness in data-constrained settings. X Hedging takes a step toward addressing these limitations by replacing neural networks with tree ensemble methods, specifically LightGBM, to estimate hedging strategies. This design provides a computationally efficient alternative while preserving strong hedging performance and moving toward more user-centered explainability. To integrate LightGBM into a reinforcement-learning-style hedging framework, we introduce an optimization approach that sequentially trains tree-ensemble models while introducing stochasticity through random training orders. In controlled experiments, X Hedging achieves performance comparable to Deep Hedging while offering greater user-centered explainability, improved computational efficiency, and substantially higher robustness when training data are limited. The numerical results demonstrate that tree-ensemble methods provide an effective computational alternative to neural networks for data-driven dynamic hedging.

cs.LG · math.OC

新论文 · v1arXiv:2611.25867

Full Rank and Stable Sampling of $\tanh$ Networks

Philipp Grohs, Matěj Trödler

原文摘要

We study the recovery of feedforward $\tanh$ networks of fixed architecture with $P$ parameters from finitely many point evaluations. First, we characterize the set of parameters where the differential of the realization map has full rank and show that it coincides with the regular parameters in the sense of Vla{č}i{ć} and B{ö}lcskei (2021). Second, we make use of this characterization to prove that almost every tuple of $2P+1$ sampling points yields uniform-norm Lipschitz stability on compact subsets of this parameter set. Third, we prove that for exact observations, projected gradient descent and projected Gauss--Newton converge locally from sufficiently close feasible initial guesses. The same evaluations also determine the realized function for arbitrary parameters and identify parameters in the characterized set up to permutations and sign changes of hidden neurons. Finally, we develop a general framework for real-analytic networks that highlights the role of full rank in stability and local reconstruction.

math.NA · cs.LG

新论文 · v1arXiv:2611.25866

From Sequence to Space: Explicit Spatial Belief with Vision-Language Distillation for Active Geo-Localization

Ziyang Xu, Peng Han, Hai Zhou

原文摘要

Active geo-localization (AGL) asks an agent to navigate a small aerial search grid and reach a target specified by an aerial image, a ground-level photograph, or text. Existing methods align the goal modalities well, but they keep the navigation state inside a large sequential Transformer and train with a dense distance reward that uses the hidden goal location. This paper proposes EAGLE, an Explicit spatial-belief Agent for Geo-Localization and Exploration. Offline training jointly optimizes a teacher-conditioned Swin branch and a context-free Swin student from the same randomly masked partial mosaics. The teacher branch attends to frozen full-scene context extracted by Qwen2.5-VL-7B, both branches receive masked-cell aerial-feature reconstruction and uncertainty supervision, while the student additionally distills stop-gradient feature predictions from the teacher. After distillation, Qwen, its cached context, and the teacher-attention branch are discarded. The retained student predicts cell-level aerial-feature embeddings and predictive uncertainty from partial observations alone. These predictions are combined with the modality-aligned goal embedding and a trajectory-derived visitation mask to construct the spatial belief tensor. The student is then frozen while a compact CNN actor-critic is optimized with PPO. EAGLE achieves state-of-the-art average success rates of 78.1%, 76.4%, and 76.8% on MASA, xBD zero-shot transfer, and the MM-GAG zero-shot aerial-goal split, respectively, without using a VLM or cached visual tokens at inference. Code will be made publicly available.

cs.RO

版本更新 · v4arXiv:2611.03043

VideoEditor-MoE: Scaling Complex Video Editing with Mixture of Experts

Xiao Cao, Fuchen Long, Yansong Qu, Wen Xiao, Xin Zhang, Xinlong Sun, Xi Chen, Zhiyong Huang

原文摘要

Complex video editing requires understanding complex instructions while jointly fulfilling multiple editing requests. Meeting these demands through dense model scaling substantially increases computational cost. We introduce VideoEditor-MoE, the first open-source mixture-of-experts framework for video editing. Beyond sparse MLP-MoE layers, MoE-Attention expands attention capacity through routed key--value experts while preserving global interactions among video tokens. Initialized from a 5B video editing model, VideoEditor-MoE scales to 14B and 56B parameters, with approximately 8B parameters activated per token in both variants. Comparisons with existing methods and two complementary baselines, Explicit-MoE and MLP-MoE, assess its benefits over explicit expert guidance and MLP-only expert architectures. Further analyses demonstrate that all experts remain active and reveal that learned routing responds more strongly to sample content and diffusion noise levels than to human-defined editing categories. Spatial analysis further indicates sensitivity to object-level semantic features. We also identify decoupled weight decay as a strategy for stabilizing expert scaling.

cs.CV · cs.AI

版本更新 · v2arXiv:2605.06404

FRInGe: Distribution-Space Integrated Gradients with Fisher--Rao Geometry

Gabriele Martino, Sebastian Tschiatschek

原文摘要

Although gradient-based attribution methods are model-faithful and scalable, a prominent variant, Integrated Gradients (IG), suffers from high sensitivity to design choices, particularly in baseline selection, path formulation, and discretization, as well as gradient saturation. To address these limitations, we propose Fisher--Rao Integrated Gradients (FRInGe), which defines both the reference and integration paths in the predictive distribution space. FRInGe replaces input baselines with a maximum-entropy predictive reference and follows a Fisher--Rao geodesic on the probability simplex. The corresponding input-space trajectory is realized through the pullback Fisher metric and stabilized by Kullback--Leibler (KL) and Euclidean trust regions; attributions are obtained by integrating input gradients along this trajectory. We further derive FRInGe-B, a variant that binarizes the predictive probability distribution, removing the computationally expensive Fisher products of FRInGe. Across six ImageNet architectures, FRInGe most clearly improves Magnitude Aligned Scoring (MAS), while remaining competitive on perturbation area-under-the-curve (AUC) metrics and infidelity.

cs.LG

版本更新 · v2arXiv:2606.11860

RePAIR: Predictive Self-Supervised Representation Learning in Chess

Christoph Koller, Johannes Fürnkranz, Timo Bertram

原文摘要

In this paper, we introduce Representation Prediction via Autoencoding using Iterative Refinement (RePAIR) - a novel self-supervised representation learning architecture that synthesizes Masked Autoencoders (MAE), Joint Embedding Predictive Architectures (JEPA), and Bidirectional Encoder Representations from Transformers (BERT). We demonstrate how it can be used to encode objects in sequential data like consecutive chess positions into compact yet meaningful representations. The basic principle of the architecture is to mask large portions of a sequence of latent states, similar to BERT and MAE. Then, we apply a lightweight Predictor to the latent representations that repairs gaps in the sequence in a lower-dimensional embedding space akin to JEPA. Our experiments in the domain of chess show that the Encoder refines the board representations such that meaningful chess concepts emerge clustered in the latent space. Furthermore, reconstructions of the masked board states show that the model is able to reason about the piece movements without relying on costly reinforcement learning methods. Lastly, we find that the resulting representation space allows for quick and intuitive dissections of chess games by observing the game path trajectories in this semantically rich space.

cs.LG