# Research Papers — 2026-07 · bundle page 130 of 222 > Full-text markdown of papers #6451–6500 of 11075 published in 2026-07, newest first. AI-readable mirrors of the full 11075-paper 2026-07 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-07/page/130 [← Previous bundle](https://aipod.dev/papers/archive/2026-07/page/129.md) [Next bundle →](https://aipod.dev/papers/archive/2026-07/page/131.md) ### IQA-T1: Tool-based Visual Evidence Reasoning for Image Quality Assessment Authors: Jinjian Wu, Jiaqi Tang, Wei Wei, Yingying Yan, Jianmin Chen, Botong Geng, Lei Zhang, Qifeng Chen Published: 2026-07-14 Categories: cs.CV, cs.AI, eess.IV arXiv: https://arxiv.org/abs/2607.12375v1 Image Quality Assessment (IQA) in open-world environments remains challenging due to limited generalization and interpretability. Recent approaches based on multimodal large language models (MLLMs) introduce textual reasoning for quality prediction, yet their judgments rely heavily on semantically biased internal representations, making them insensitive to low-level perceptual degradations. We propose IQA-T1, a tool-based visual evidence reasoning framework that augments MLLM reasoning with explicit perceptual observations. During inference, the model autonomously invokes specialized analysis tools to generate structured visual evidence, such as noise residual maps, gradient statistics, and frequency spectra, which are progressively integrated into the reasoning process. To support this paradigm, we construct Q-Tool, a dataset containing 11k multimodal reasoning chains grounded in tool-generated evidence. Extensive experiments on seven IQA benchmarks show that IQA-T1 achieves the best overall performance across datasets while producing interpretable and evidence-grounded quality assessments. Code and dataset are available at https://github.com/zibuyu-02/IQA-T1. --- ### UMSS: Towards Unsupervised Multi-modal Semantic Segmentation Authors: Haitian Zhang, Thai Duy Nguyen, Xiangyuan Wang, Mohan Liu, Lin Wang Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.12372v1 Multimodal semantic segmentation (MSS) is essential for robust perception in complex environments, yet its potential remains largely untapped because of the prohibitive cost of human annotations. While unsupervised semantic segmentation (USS) has achieved strong results on a single RGB modality, its naive extension to multimodal data is often hindered by fusion degradation. This occurs because, without explicit supervision, existing frameworks struggle to reconcile the heterogeneous structural patterns captured by different sensors and therefore fail to effectively exploit their complementary information. In this paper, we make the first attempt to address the novel problem of Unsupervised Multimodal Semantic Segmentation (UMSS), aiming to effectively exploit complementary sensor information in a fully label free setting. To this end, we propose UniM2 (Unified Multimodal), a novel framework built on DINOv3 that transforms conventional fusion methods into consistent performance gains. Our key idea is to learn a unified latent space driven by Cross Modal Correspondence Synergy (CMCS) to extract intrinsic shared semantic cues, bypassing the need for label guided adaptive fusion. To mitigate inherent intermodal conflicts, we introduce a Cross Modal Harmonizer (CMH) that designates RGB as a stable reference, effectively suppressing inconsistent relational supervision while guiding the model to exploit complementary structural features. Extensive experimental results on NYU Depth v2 and MFNet show that UniM2 improves mIoU by 6.4% and 9.8%, respectively, demonstrating clear advantages over existing frameworks for UMSS. --- ### Lost in Visual Translation: A VLM-Assisted Perceptual-Semantic Coherence Framework for EEG-to-Image Reconstruction Authors: Sukriti Tiwari, BHVSP Subrahmanyam, Nidhi Goyal, Sai Amrit Patnaik Published: 2026-07-14 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2607.12364v1 EEG-to-image evaluation should distinguish visual fidelity from recoverable meaning. Yet EEG-derived reconstructions are blurry, distorted, and low-detail, causing SSIM, LPIPS, and CLIP to penalize semantically recoverable outputs or reward plausible but incorrect ones. We analyze 6,855 ground-truth/reconstruction pairs from ATM, ENIGMA, BrainVis, and DreamDiffusion using semantic probes, caption harshness and blind-spot rates, and controlled degradations. Pixel metrics show near-zero correlation with semantic consistency, while representation metrics conflate perceptual and semantic errors. We therefore introduce a BCI-aware framework in which four VLMs assess image pairs through structured questions, producing Tolerant Perceptual Alignment Scores (T-PAS) and Tolerant Semantic Alignment Scores (T-SAS). Their consensus is distilled into the BCI-Coherence Score (BCS), a compact evaluator achieving a T-PAS MAE of 0.079 (r = 0.700) and a T-SAS MAE of 0.082 (r = 0.850) on our data. Human validation shows highly reliable joint coherence judgments, with Cohen's kappa = 0.882 +/- 0.174 and Krippendorff's alpha = 0.882, supporting perceptual-semantic recoverability over generic visual similarity. Code and resources are available at https://sukt03.github.io/BCS/. --- ### Implicit 4D Gaussian Splatting for Fast Motion with Large Inter-Frame Displacements Authors: Seung-gyeom Kim, Areum Kim, Yongjae Yoo, Sukmin Yun Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.12362v1 Recent 4D Gaussian Splatting (4DGS) methods often fail under fast motion with large inter-frame displacements, where Gaussian attributes are poorly learned during training, and fast-moving objects are often lost from the reconstruction. In this work, we introduce Spatiotemporal Position Implicit Network for 4DGS, coined SPIN-4DGS, which learns Gaussian attributes from explicitly collected spatiotemporal positions rather than modeling temporal displacements, thereby enabling more faithful splatting under fast motions with large inter-frame displacements. To avoid the heavy memory overhead of explicitly optimizing attributes across all spatiotemporal positions, we instead predict them with a lightweight feed-forward network trained under a rasterization-based reconstruction loss. Consequently, SPIN-4DGS learns shared representations across Gaussians, effectively capturing spatiotemporal consistency and enabling stable high-quality Gaussian splatting even under challenging motions. Across extensive experiments, SPIN-4DGS consistently achieves higher fidelity under large displacements, with clear improvements in PSNR and SSIM on challenging sports scenes from the CMU Panoptic dataset. For example, SPIN-4DGS notably outperforms the strongest baseline, D3DGS, by achieving +1.83 higher PSNR on the Basketball scene. --- ### Same Loss, Same Noise, Opposite Schedules: Noise Structure and Optimizer Normalization Jointly Determine Whether Learning-Rate Cooldown Helps Authors: Subham Singh, Ashutosh Mishra, Subha Raut Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.12360v1 The cooldown phase of a warmup-stable-decay (WSD) learning-rate schedule, now a default in large-model pretraining, lowers the final training loss in some settings and does nothing in others. We give a provable account of which case obtains, and it turns on two properties together: the structure of the gradient noise and whether the optimizer normalizes its update. On a strongly convex objective with multiplicative (gradient-proportional) noise, stochastic gradient descent contracts geometrically at a constant learning rate, so cooldown has nothing to improve. Under the same objective and noise, sign-based and normalized methods, the standard surrogates for adaptive optimizers, settle on a noise floor of order $η^2$ and reach the minimizer only as the learning rate is driven to zero; any additive noise then reinstates a floor for every method. The mechanism is elementary: an SGD step shrinks in proportion to the gradient and so anneals itself, whereas a normalized step keeps unit scale and cannot. We solve the signSGD stationary law on the quadratic exactly and obtain the floor constant in closed form, prove a local form of the dissociation under $(L_0,L_1)$-smoothness, extend the floor to normalized SGD in dimension d>1 by a scale-invariance argument, and establish robustness to momentum and heavy-tailed noise. Simulation confirms every prediction, and we demonstrate the resulting noise-regime diagnostic on a real classification task with directly measured gradient noise. The mechanism explains whether cooldown helps; the interior cooldown fraction used at scale lies outside stationary landscape-and-noise geometry. --- ### One Threshold Is Not Enough: Prompt-Invariant Caching Schedules for Video Diffusion Authors: Om Agrawal, Saurabh Agarwal, Aditya Akella Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.12358v3 Video diffusion models remain expensive because denoising repeatedly evaluates a costly model backbone. Dynamic caching methods like TeaCache, EasyCache, and DiCache reduce this cost by reusing prior outputs while an accumulated drift signal remains below a threshold $τ$. Although existing methods improve the drift signal, they hold $τ$ fixed throughout denoising. We show that fixed thresholds are structurally suboptimal: varying $τ$ across denoising achieves quality-latency tradeoffs unattainable by any fixed threshold across three caching methods with distinct drift signals. We further show that, surprisingly, each method's drift trajectory is nearly prompt-invariant across seven method-model combinations. The shape of the drift trajectory is determined by the model and caching method; therefore, a threshold schedule can be calibrated offline and reused across generations. Based on these findings, we present ACID (Adaptive Caching for vIDeo generation), which uses a low threshold during critical regions where drift changes rapidly and a high threshold across stable regions. ACID identifies these regions offline from the drift signal's second derivative, requires no training, and adds negligible runtime overhead. Across TeaCache, EasyCache, and DiCache on HunyuanVideo, Wan 2.1, and CogVideoX, ACID pushes the speed-quality Pareto frontier beyond fixed thresholds. On TeaCache with HunyuanVideo, it achieves 2.16x speedup over no caching and 38% additional speedup over a conservative fixed threshold, with less than 0.3 dB PSNR, 0.01 SSIM, and 0.01 LPIPS degradation relative to that fixed-threshold configuration. --- ### ACID: Adaptive Caching for vIDeo generation Authors: Om Agrawal, Saurabh Agarwal, Aditya Akella Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.12358v2 Video diffusion models produce high-quality generations but remain slow at inference due to their sequential denoising procedure. Caching-based acceleration methods address this by reusing intermediate model outputs: leading dynamic approaches such as TeaCache, EasyCache, and DiCache accumulate a drift signal and skip expensive model evaluations when accumulated drift stays below a fixed threshold $τ$. This threshold controls an apparent tradeoff - raising it yields faster generation at the cost of visual quality, while lowering it preserves quality but sacrifices speed. We show this tradeoff is not fundamental; it is an artifact of holding $τ$ constant throughout denoising. We identify the existence of critical steps - timesteps where the drift signal changes rapidly - and show that applying a low threshold selectively at these steps while caching aggressively elsewhere recovers most of the quality of conservative caching at substantially higher inference speeds. Building on this insight, we propose ACID, a lightweight, training-free wrapper that monitors the rate of change of each method's existing drift signal to dynamically switch between a low and a high threshold. ACID is signal-agnostic and modular: it requires no retraining and plugs directly into existing dynamic caching methods without modifying their core mechanisms. Evaluated across three caching methods (TeaCache, EasyCache, DiCache) and three open-source video diffusion models (HunyuanVideo, Wan 2.1, CogVideoX), ACID consistently expands the Pareto frontier of visual quality versus inference speed beyond what any fixed threshold achieves. In particular, on TeaCache and HunyuanVideo, ACID achieves up to 2.16x speedup over the no-caching baseline, and up to 38% additional speedup over the conservative fixed-threshold baseline with negligible (<0.3 dB PSNR, <0.01 SSIM, <0.01 LPIPS) quality degradation. --- ### Reducing information dependency does not cause training data privacy. Adversarially non-robust features do Authors: Rasmus Torp, Shailen K. Smith, Adam Breuer Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.12354v2 In this paper, we challenge the prevailing view that information dependency (including rote memorization) drives training data exposure to image reconstruction attacks. We show that extensive exposure can persist without rote memorization and is instead caused by a tunable connection to adversarial robustness. We begin by presenting three surprising results: (1) recent defenses that inhibit reconstruction by Model Inversion Attacks (MIAs), which evaluate leakage under an idealized attacker, do not reduce standard measures of information dependency (HSIC); (2) models that maximally memorize their training datasets remain robust to MIA reconstruction; and (3) models trained without seeing 97% of the training pixels, where recent information-theoretic bounds give arbitrarily strong privacy guarantees under standard assumptions, can still be devastatingly reconstructed by MIA. To explain these findings, we provide causal evidence that privacy under MIA arises from what the adversarial examples literature calls ``non-robust'' features (generalizable but imperceptible and unstable features). We further show that recent MIA defenses obtain their privacy improvements by unintentionally shifting models toward such features. To establish this causal relationship, we introduce Anti Adversarial Training (AT-AT), a training regime that intentionally learns non-robust features to obtain both superior reconstruction defense and higher accuracy than state-of-the-art defenses. Our results revise the prevailing understanding of training data exposure and reveal a new privacy-robustness tradeoff. --- ### Reducing information dependency does not cause training data privacy. Adversarially non-robust features do Authors: Rasmus Torp, Shailen K. Smith, Adam Breuer Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.12354v1 In this paper, we challenge the prevailing view that information dependency (including rote memorization) drives training data exposure to image reconstruction attacks. We show that extensive exposure can persist without rote memorization and is instead caused by a tunable connection to adversarial robustness. We begin by presenting three surprising results: (1) recent defenses that inhibit reconstruction by Model Inversion Attacks (MIAs), which evaluate leakage under an idealized attacker, do not reduce standard measures of information dependency (HSIC); (2) models that maximally memorize their training datasets remain robust to MIA reconstruction; and (3) models trained without seeing 97% of the training pixels, where recent information-theoretic bounds give arbitrarily strong privacy guarantees under standard assumptions, can still be devastatingly reconstructed by MIA. To explain these findings, we provide causal evidence that privacy under MIA arises from what the adversarial examples literature calls ``non-robust'' features (generalizable but imperceptible and unstable features). We further show that recent MIA defenses obtain their privacy improvements by unintentionally shifting models toward such features. To establish this causal relationship, we introduce Anti Adversarial Training (AT-AT), a training regime that intentionally learns non-robust features to obtain both superior reconstruction defense and higher accuracy than state-of-the-art defenses. Our results revise the prevailing understanding of training data exposure and reveal a new privacy-robustness tradeoff. --- ### Filtering-out poor-quality images for data preparation Authors: Roopdeep Kaur, Gour Karmakar, Muhammad Imran Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.12352v1 Filtering noise is a fundamental part of data preparation that enhances image quality for applications such as object segmentation, detection, and recognition. Various noise reduction techniques are proposed in the literature, including the use of median, Gaussian, and bilateral filters. Convolutional neural networks (CNNs) have gained popularity in image denoising owing to their ability to extract complex patterns and features from data. CNNs are highly adaptable, making them effective tools for various image-denoising tasks. One drawback of CNN-based techniques is that they require an appropriate training dataset and all images to be resized. Another notable drawback of all these filtering techniques is that they work for certain types of environmental and camera noises. To bridge this research gap, in this paper, for the first time, instead of denoising, we propose an approach that filters out poor-quality images for various environmental and camera impacts. In our approach, quality is assessed using an image quality assessment metric and an optimum threshold is used to filter out poor-quality images. We also ensure that a sufficient number of images remain to develop the deep learning (DL) model. The results produced using real and simulated traffic and object recognition data demonstrate the performance supremacy of the proposed approach compared with the state-of-the-art approaches. The average recognition accuracy for our proposed approach is 93.8% for the traffic sign recognition dataset and 84.9% for the object recognition dataset. This indicates our model's potential for real-life applications such as autonomous vehicles. --- ### Generating Developable 3D Molecules via Pocket-Conditioned Diffusion and Property-Aware Optimization Authors: Ruoxi Gao, Jiangweizhi Peng, Ziqi Chen, Frazier N. Baker, David C. Kombo, John L. Kane, Andrew A. Scholte, Yi Li, Matthew J. LaMarche, Luigi I. Iconaru, Hans-Peter Biemann, Mingyi Hong et al. Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.12349v1 Drug discovery and development is time-consuming and resource-intensive, motivating computational approaches such as diffusion models for de novo drug design. Many such models follow the structure-based drug design (SBDD) paradigm, generating molecules to fit a target binding pocket. However, existing diffusion-based SBDD methods typically couple pocket and ligand representation learning, model interactions only at the atom level, and prioritize binding affinity over other developability properties. Here, we introduce conDitar-dev, a conditional diffusion-based SBDD framework for generating ligands with strong binding affinities and favorable ADMET properties. It consists of three modules: msPRL, a pretrained multi-scale pocket representation learning module; conDitar, a pocket-conditioned diffusion model guided by msPRL representations; and paOPT, a generation-time method for optimizing ligand developability. On a newly curated benchmark of human disease targets, conDitar outperforms state-of-the-art SBDD baselines, achieving an average binding score of -8.85 kcal/mol. Across five ADMET properties, conDitar-dev improves performance by up to 73% over conDitar. To further validate the abilities of conDitar-dev to generate developable molecules, we have applied it to two validated druggable targets: programmed death-ligand 1 (PD-L1) and colony-stimulating factor 1 receptor (CSF1R) proteins. Top-ranked generatively designed molecules and their analogs have been experimentally synthesized and biologically tested. Two molecules generated directly by conDitar-dev for PD-L1 exhibited SPR-derived $K_D$ values of 3.49 and 3.75 $μ$M, respectively. Hit expansion based on conDitar-dev-designed molecules identified selective CSF1R inhibitors with IC$_{50}$ values as low as 200 nM, while also uncovering opportunities for drug repositioning. --- ### Forecasting Inflation with Microdata: An Adaptive Machine Learning Approach Authors: Catherine Chen, Chen Gao, Jonathon Hazell, Lihua Lei, Chen Lian Published: 2026-07-14 Categories: econ.GN, stat.ME, stat.ML arXiv: https://arxiv.org/abs/2607.12345v1 Does microeconomic heterogeneity help to forecast aggregate inflation in a non-stationary environment? We develop a scan test for whether one forecast outperforms another, over an interval with unknown starting point and duration. To exploit any occasional forecasting power that the scan test detects, we design an adaptive machine learning pipeline. We encode the distribution of price changes into a high-dimensional vector, which we combine with a gradient boosted trees algorithm. We then combine this micro forecast with other benchmark forecasts, using an adaptive algorithm that makes use of the micro forecast only when it performs well. We apply the pipeline to UK microdata, with four main results. First, the micro forecast outperforms a univariate benchmark, but only in the volatile period after 2020. Second, the scan test detects periods of micro outperformance, so the micro forecast enters the combined forecast. Third, the combined forecast performs comparably to the univariate benchmark before 2020 and better at every horizon after 2020. Fourth, the value of microdata for the combined forecast materializes after 2020. We conclude that microdata are valuable for forecasting aggregate inflation, but only after large shocks. --- ### Policy-Conditioned Constrained Decoding for Column-Level Access Control in Text-to-SQL Authors: Ryoto Miyamoto, Xin Fan, Hayato Yamana Published: 2026-07-14 Categories: cs.CL, cs.CR, cs.DB arXiv: https://arxiv.org/abs/2607.12341v2 Text-to-SQL is increasingly deployed across trust boundaries between data providers and users. Such deployment must balance three competing requirements: policy compliance, answer coverage, and bounded cost. Existing approaches typically decide refusal based on which columns a query mentions and enforce it stochastically. Whether a query is compliant, however, depends not only on which columns appear but on how they are used, and stochastic enforcement cannot deterministically rule out violations. We formalize this requirement as a column-use policy over semantic use: output, filter condition, and aggregation argument. We integrate the policy by aligning each role with grammar productions tracked by the decoder. The resulting system, PCC-SQL, applies a per-token logits mask that deterministically eliminates single-query column-use violations on the supported SQL fragment in a single decoding pass. Across three benchmarks and three open-source models, PCC-SQL achieves 0% Leakage Rate and Coverage up to 88.7% on Spider-CU, while staying within +10% tokens of direct prompting. We additionally assess semantic alignment with execution accuracy. --- ### Policy-Conditioned Constrained Decoding for Column-Level Access Control in Text-to-SQL Authors: Ryoto Miyamoto, Xin Fan, Hayato Yamana Published: 2026-07-14 Categories: cs.CL, cs.CR, cs.DB arXiv: https://arxiv.org/abs/2607.12341v1 Text-to-SQL is increasingly deployed across trust boundaries between data providers and users. Such deployment must balance three competing requirements: policy compliance, answer coverage, and bounded cost. Existing approaches typically decide refusal based on which columns a query mentions and enforce it stochastically. Whether a query is compliant, however, depends not only on which columns appear but on how they are used, and stochastic enforcement cannot deterministically rule out violations. We formalize this requirement as a column-use policy over semantic use: output, filter condition, and aggregation argument. We integrate the policy by aligning each role with grammar productions tracked by the decoder. The resulting system, PCC-SQL, applies a per-token logits mask that deterministically eliminates single-query column-use violations on the supported SQL fragment in a single decoding pass. Across three benchmarks and three open-source models, PCC-SQL achieves 0% Leakage Rate and Coverage up to 88.7% on Spider-CU, while staying within +10% tokens of direct prompting. We additionally assess semantic alignment with execution accuracy. --- ### How Many Tasks Are Enough for Agent Benchmark Decisions? A Replay Analysis of Public LLM Agent Benchmarks Authors: Wei-Jung Huang Published: 2026-07-14 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.12338v1 Agent benchmarks often compare two agents after all tasks have run, but costly evaluations make partial runs tempting. A task fraction alone does not show whether a partial run supports the same pairwise conclusion as the completed benchmark. We study this question by replaying completed public task-level records from SWE-bench, AppWorld, and tau-bench. A partial budget counts as enough only when it supports the completed benchmark's decision, covers required task groups, and leaves no more than a target fraction of comparisons unresolved. The required task fraction varies sharply. At the strict 0 percentage point threshold on a 5 percentage point budget grid, AppWorld first meets all targets at 15 percent, tau-bench at 25 percent, and SWE-bench Verified at 90 percent; SWE-bench Lite does not meet all targets by 95 percent under the primary coverage rule. Partial-evaluation reports should state how much one agent must outperform another, how tasks are selected, what coverage rule is required, what decision rule is used, and how many comparisons may remain unresolved. --- ### Robust Design of Integrated Sensing and Communication in LEO Satellite Systems Authors: Hezhen Yang, Xiaoming Chen, Qi Wang Published: 2026-07-14 Categories: cs.IT, cs.LG arXiv: https://arxiv.org/abs/2607.12337v1 With the growing demand for satellite sensing and communication, the limited wireless resources are difficult to support multiple satellite systems. Therefore, it is desired to investigate integrated sensing and communication (ISAC) in low Earth orbit (LEO) satellite systems to enable multi-functionality within a single satellite, thereby saving both spectrum and orbital resources. In this paper, a framework for ISAC in LEO satellite systems is established, where a satellite can simultaneously sense multiple targets and serve multiple communication users (CUs) over the same spectrum. Considering the limited onboard energy of satellite, a novel robust beamforming design algorithm is developed with the goal of minimizing total transmit power while satisfying the mean squared error (MSE) requirements for sensing and signal-to-interference-plus-noise ratio (SINR) requirements for communication in presence of channel phase uncertainty which exacerbates the cross-functional interference. According to theoretical analysis, the proposed algorithm for ISAC in LEO satellite systems is effective. Moreover, extensive simulations confirm the superiority of the proposed algorithm over baselines. --- ### Evaluating Health Misinformation in Low-Resource Languages: Integrating Small Language Models with a Culturally-Sensitive Responsible NLP Framework (Bangla as a Case Study) Authors: Farnaz Farid, Raihan Alam, Al Al-Areqi, Farhad Ahamed, Muhammad Hassan Khan, Sadia Hossain, Irena Veljanova, Anika Tabassum Binte Hossain Published: 2026-07-14 Categories: cs.CL, cs.AI, cs.CY, cs.ET, cs.HC arXiv: https://arxiv.org/abs/2607.12336v1 Artificial Intelligence (AI) technologies, while serving as a foundational enabler for modern social media and digital health services, exert a bivalent effect by simultaneously acting as a combatant against and a spread vector for misinformation. A prevalent challenge in mitigating this issue arises in non-English contexts and low socioeconomic classes, where limited data hinders the training of AI models for effective detection. Consequently, culturally and linguistically diverse (CALD) communities struggle to access trustworthy health information through AI-driven tools. Current AI tools underperform due to a lack of training data and are largely unable to consider language nuances and traditions in non-English contexts. This research addresses these gaps by proposing a CALD-friendly AI-based health misinformation detector and providing a dashboard for medical professionals to analyse this misinformation, a critical step toward mitigating a growing concern among CALD populations. To this end, we conduct a series of experiments using a Bangla-translated health misinformation dataset to evaluate the performance of various Small Language Models (SLMs). SLMs are particularly relevant in this context given the frequent underperformance of Large Language Models (LLMs), which often stems from insufficient domain-specific knowledge and the prohibitive costs of resource-intensive fine-tuning. The results demonstrate that Phi-4 is the superior model, achieving an ideal balance between precision and recall in claim extraction. Then, to mitigate the limitations of SLMs, we design and test a novel health misinformation detection framework grounded in Responsible Natural Language Processing (NLP), which incorporates cultural sensitivity, potential for harm, and communication quality, thereby providing a holistic lens for evaluating misinformation in low-resource languages. --- ### ProtoPointNet: Prototype-Based Interpretable Classification of 3D Dental Point Clouds with Verifiable Spatial Activations Authors: George V. Jose, Thao Liang Chiam, Toby Hughes, Dilan Patel, Alan Brook, Lyle J. Palmer, Nikhil Cherian Kurian Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.12335v1 Prototype-based networks provide inherently interpretable classification by linking predictions to learned exemplars, but their use in 3D point clouds and clinical surface-pair reasoning remains limited. We introduce ProtoPointNet, a prototype-based model for dental occlusion classification from registered upper--lower intraoral arch pairs. Each point is encoded by a 14-dimensional descriptor combining local surface geometry, curvature, and explicit inter-arch displacement and clearance, exposing occlusal relationships to prototype matching. A shared multi-task point-cloud backbone learns axis-specific prototype heads for sagittal-left, sagittal-right, vertical, transverse, and midline classification. To support limited clinical data, we train prototypes from scratch using auxiliary supervision and encoder-freeze hand-off. On Bits2Bites, ProtoPointNet achieves mean test macro-F1 of 0.724 and AUROC of 0.825, with strongest performance on vertical (F1 0.828) and sagittal-left classification (F1 0.807). Projected prototype activations localise to anatomically plausible regions, including posterior molars and premolars for cross-bite evidence and anterior incisors for bite-depth evidence. These results support prototype-based reasoning as a transparent, spatially grounded alternative to black-box 3D classifiers for dental surface-pair analysis. --- ### QUBO-Optimized Evidence Selection for Retrieval-Augmented Question Answering with Unconventional Solvers Authors: Rahul Singh, Madhav Vadlamani Published: 2026-07-14 Categories: cs.CL, cs.ET arXiv: https://arxiv.org/abs/2607.12334v1 Retrieval-augmented question answering depends on selecting evidence passages that jointly support answer generation. However, many RAG pipelines rely on top-\(k\) ranking, where passages are selected mainly by individual relevance scores, even though multi-hop questions often require complementary evidence satisfying multiple information requirements. Recent LLM-based selectors address this by treating retrieval as set selection, but using an LLM for this intermediate stage can be costly and difficult to scale. In this work, we formulate evidence selection as a Quadratic Unconstrained Binary Optimization (QUBO) problem. Given a question, candidate passages, and decomposed information requirements, our method constructs an energy function that balances relevance, requirement coverage, support strength, redundancy, complementarity, and compactness. Low-energy solutions correspond to compact evidence subsets that cover the needed requirements while avoiding unnecessary or repetitive context. The selected passages are then passed to a downstream language model for answer generation, separating combinatorial evidence selection from semantic answer generation. We evaluate the proposed QUBO selector on HotpotQA and compare it with LLM-based set selectors and non-LLM baselines including BM25, relevance top-\(k\), maximal marginal relevance, hybrid lexical--semantic ranking, greedy coverage, and random selection. The QUBO selector achieves competitive exact-match and token-F1 performance relative to LLM-based selectors while providing a solver-compatible formulation for structured evidence selection. These results suggest that multi-hop evidence selection can be cast as discrete optimization, opening a path toward RAG pipelines where LLMs are reserved for semantic processing and answer generation, while context selection is handled by Ising/QUBO-compatible solvers. --- ### Gradient Flow Dynamics and Implicit Bias of Diagonal Linear Networks under Infinitesimal Initialization Authors: Jiajie Zhao, Jianxing Wang, Junjie Yang, Zhiwei Bai, Yaoyu Zhang Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.12332v1 We study the gradient flow dynamics of diagonal linear networks for regression tasks under infinitesimal initialization. Extending Theorem 1 from Pesme & Flammarion (2023), we generalize the analysis to both deep diagonal linear networks and a broader class of two-layer diagonal linear networks (as defined in Definition 4.1). Specifically, we demonstrate that the training trajectories of these models can be equivalently characterized by the proposed Algorithm 1. We further prove that this algorithm converges to the solution of a modified $ \mathcal{l}_1 $ norm minimization problem. As a result, we establish that the implicit bias of both network architectures corresponds to a modified $ \mathcal{l}_1 $ norm in the regime of infinitesimal initialization. Additionally, we provide insights into the underlying mechanisms governing these dynamics by identifying the Structural Invariant Manifold (SIM) (Zhao et al., 2026) as the key geometric structure that shapes the learning process. --- ### DM-KG: A Novel Method for Boosting Spatial Cognition of Vision-Language Models in Street View Imagery Authors: Xinyue Xu, Zheng Zhang, Kunyang Ma, Ge Zhu, Lianshuai Cao, Lei Wang, Zixuan Li, Yi Cheng Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.12319v1 As vision-language models (VLMs) are increasingly deployed in geospatial question answering and visual scene understanding, improving their spatial cognition capability on street view imagery for complex logical reasoning has emerged as a key research priority. However, existing VLMs frequently suffer from "spatial semantic hallucinations" when perceiving object locations, distances, and directions in real-world street view scenes. Furthermore, such errors are often recalcitrant to tracing and calibration, posing a critical bottleneck for their practical deployment in geospatial tasks. To address this pressing challenge, this study proposes DM-KG (Direction-Metric Knowledge Graph), a structurally grounded spatial representation framework for street view imagery. By explicitly extracting directional and metric relationships between entities from a single 2D image, this framework enhances the spatial reasoning accuracy of VLMs through a structured knowledge graph. Specifically, we integrate panoptic segmentation with metric depth estimation to robustly compute entity-level 3D spatial coordinates. Subsequently, we encode the clock azimuths and Euclidean distances of entity pairs into a JSON-formatted knowledge graph, which is injected into the VLM as an explicit geometric prior to guide spatial reasoning. Experimental results on public spatial question-answering (QA) benchmarks demonstrate that DM-KG reduces the mean absolute error (MAE) in distance estimation by 31.1% and the mean angular error in direction judgment by 65.8%, while simultaneously maintaining a high QA success rate. By establishing a complete, augmented reasoning pipeline, this research significantly improves the spatial cognitive capabilities of VLMs in street view scenarios, thereby providing a flexible, generalized, and interpretable framework for geographic visual question answering (GeoVQA) in open environments. --- ### LakeQuest: A Three-Domain Benchmark for Grounded Question Answering across Data Lakes Authors: Michael Solodko, Steven Gong, Guangwei Yu, Satya Krishna Gorti, Jesse C. Cresswell, Victor Zhong Published: 2026-07-14 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2607.12310v3 While modern question answering (QA) systems excel on clean, schema-aligned corpora, real-world knowledge is rarely so neatly packaged. Answering questions over enterprise and scientific data lakes requires systems to navigate heterogeneous, weakly structured collections of tables, passages, and linked metadata. Current benchmarks abstract away this noisy discovery process, failing to evaluate end-to-end performance. To bridge this gap, we introduce LakeQuest, a human-validated benchmark of 9,846 QA pairs designed to evaluate the end-to-end retrieve-and-synthesize pipeline over realistic data lakes. LakeQuest spans three diverse domains (AI/ML metadata, retail banking, and multimodal biomedical drug information) and pairs every question with exact, modality-aware evidence pointers. By isolating source discovery from cross-modal synthesis, LakeQuest exposes critical failure modes in modern QA systems. Our baseline evaluations, including standard Retrieval-Augmented Generation (RAG) and agentic tool-use methods, reveal that high-quality retrieval does not guarantee correct reasoning. Systems consistently struggle with relation chaining in metadata graphs, policy grounding in bank ledgers, and joint tabular QA in biomedical contexts, highlighting the need for robust discovery and faithful cross-file composition mechanisms in future agentic QA systems. --- ### LakeQuest: A Three-Domain Benchmark for Grounded Question Answering across Data Lakes Authors: Michael Solodko, Steven Gong, Guangwei Yu, Satya Krishna Gorti, Jesse C. Cresswell, Victor Zhong Published: 2026-07-14 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2607.12310v2 While modern question answering (QA) systems excel on clean, schema-aligned corpora, real-world knowledge is rarely so neatly packaged. Answering questions over enterprise and scientific data lakes requires systems to navigate heterogeneous, weakly structured collections of tables, passages, and linked metadata. Current benchmarks abstract away this noisy discovery process, failing to evaluate end-to-end performance. To bridge this gap, we introduce LakeQuest, a human-validated benchmark of 9,846 QA pairs designed to evaluate the end-to-end retrieve-and-synthesize pipeline over realistic data lakes. LakeQuest spans three diverse domains (AI/ML metadata, retail banking, and multimodal biomedical drug information) and pairs every question with exact, modality-aware evidence pointers. By isolating source discovery from cross-modal synthesis, LakeQuest exposes critical failure modes in modern QA systems. Our baseline evaluations, including standard Retrieval-Augmented Generation (RAG) and agentic tool-use methods, reveal that high-quality retrieval does not guarantee correct reasoning. Systems consistently struggle with relation chaining in metadata graphs, policy grounding in bank ledgers, and joint tabular QA in biomedical contexts, highlighting the need for robust discovery and faithful cross-file composition mechanisms in future agentic QA systems. --- ### What Does a Temporal Benchmark Score Measure? Decomposing Channel Use in Video VLM Evaluation Authors: Farrukh Rahman Published: 2026-07-14 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2607.12304v1 A score on a temporal video question answering benchmark is meant to measure that a model has temporal understanding, but it conflates two questions. 1. The task question: is the question even temporal, does it need several frames and their order? and 2. The channel question, when it does, does the model recover the order from the pixels, or read it off the positional encoding (RoPE)? Most of a temporal score answers neither, a single frame and answer priors often carry it. The field's validity checks, frame-shuffle sensitivity and the accuracy gained from the full video, speak only to the task question. We contribute a label-free screen for the channel question, the reversal-drop: the accuracy lost when the visual sequence is reversed while RoPE remains forward. It can be applied to compatible temporal benchmarks without new annotations. Paired reverse labels, or tasks whose labels transform deterministically under reversal, distinguish models that follow reversed content from those merely disrupted by the conflict. Molmo2 answers the forward event reading order off positions, while Qwen3-VL answers the reversed event it actually sees, reading visual order (comparatively). We call them position-dominant and visual-sequence-dominant. The split holds across two benchmarks and several temporal tasks at two scales, and activation patching shows it is a real internal property, not an artifact of the conflict. The distinction matters, the two channels fail on opposite inputs so two models with similar score are not interchangable, i.e. an aggregate score does not reflect potential failure modes. --- ### MobileSAM2: Lightweight Segment Anything for Spatial Intelligence Authors: Kai Jiang, Jiaxing Huang, Jingyi Zhang, Weiying Xie, Yunsong Li, Yufei Wang, Aoran Xiao, Dacheng Tao Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.12297v1 The recent large video foundation model, SAM2, enables segment anything in both images and videos, serving as a powerful base model for various applications. However, many of such use cases require to operate on resource-constrained devices like mobile phones and laptops. In this work, we aim to make SAM2 more mobile-friendly by distilling the heavyweight SAM2 into a lightweight model, facilitating segment anything in both images and videos on mobile devices. To this end, we propose Hypergraphical Knowledge Distill (HyperKD), which introduces the idea of hypergraph into knowledge distillation, aiming to effectively model and transfer SAM2's generalizable and comprehensive knowledge. HyperKD consists of Temporal HyperKD and Granularity HyperKD that construct hypergraphs to explicitly model and extract the generalizable temporal knowledge and the comprehensive multi-granularity knowledge from SAM2 respectively, which are then distilled into the lightweight student model by aligning it with the constructed hypergraphs. Besides, we present MobileSAM2, a new family of lightweight SAM2 that balances efficiency and effectiveness via searching the best model architectures with HyperKD during model size reduction. Extensive experiments validate MobileSAM2 across multiple benchmarks and show promising generalization performance on embodied AI tasks. --- ### A Comparative Analysis of Institutional and Course Generative AI Policies within Higher Education: Implications for Instruction in Computing Education Authors: Amrita Ganguly, Aditya Johri, Nora McDonald, Areej Ali, Umama Dewan, Aayushi Hingle Collier Published: 2026-07-14 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2607.12296v1 With the increased use of generative AI (GenAI) applications such as ChatGPT, higher education institutions (HEIs) have released a range of guidelines and policies to direct adoption within their institutions. In computer science (CS) courses GenAI adoption is especially high and the implications for student learning are significant. At the same time, instructors have also been forced to address the use of GenAI as students have started to use it for a range of functions. Currently, comparative analysis of guidance provided by institutions and its uptake in instruction is lacking. In this paper we bridge this gap by comparing institutional and computing course level guidance to better understand this terrain. We utilize secondary analysis of institutional and course syllabi guidelines from higher education institutions in the U.S. classified as research-intensive. Our findings reveal that although institutional guidance is more pro-use, at the course-level the uptake is still guarded. We discuss the implications and propose an instructor-centered framework to guide future adoption of GenAI. --- ### A Longitudinal Analysis of Public Discourse on AI Ethics in Education Using Twitter Data Authors: Akriti Bagale, Nafisa Mehjabin, Ali Ünlü, Aditya Johri Published: 2026-07-14 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2607.12295v1 The rapid integration of artificial intelligence (AI) and generative AI (GenAI) into education presents significant opportunities to enhance teaching and learning, while raising ethical concerns about the responsible use of these technologies in educational settings. Understanding how the public perceives and debates these issues is increasingly important for educators, institutions, and policymakers seeking to integrate AI responsibly and equitably. Social media platforms, where such debates unfold frequently and at scale, offer a valuable lens for capturing large-scale, real-time public reactions to key developments as they emerge. In this study, we analyse five years (2019-2024) of discourse on Twitter (now X) to trace the evolving public conversation around AI ethics in education, paying particular attention to the release of ChatGPT as a pivotal moment that reshaped the nature and tone of that discourse. Using BERT-based topic modelling and SetFit sentiment analysis to identify dominant themes and track sentiment over time, we find that the discourse has been predominantly positive across the observation period, with negative sentiment concentrated around specific ethical controversies. More recently, anxieties about academic integrity and the broader implications of generative AI have come to dominate the conversation. Rather than reflecting a polarized debate, public discourse appears pragmatic and largely receptive to AI integration, though accompanied by growing calls for ethical oversight and institutional accountability. By providing a longitudinal account of public sentiment surrounding AI ethics in education, this study informs educators, institutions, and policymakers an empirically grounded understanding of public expectations, informing the development of responsible, transparent, and equitable approaches to AI integration across educational contexts. --- ### Adaptive Cross-Modal Fusion with Sparse Attention for Pedestrian Crossing Intention Prediction Authors: Md Mahfuzur Rahman, Pengzhan Zhou, A F M Abdun Noor, Md Imam Ahasan, Kah Ong Michael Goh, S. M. Hasan Mahmud, Md Mustafizur Rahman, Kaixin Gao Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.12293v1 Predicting pedestrian crossing intention is a safety-critical task for autonomous driving, yet existing approaches often rely on single-modal inputs or dense multimodal fusion strategies that inadequately capture complementary visual and kinematic information while introducing redundant inter-modal interactions. We propose ADAPT (Adaptive Domain-Aware Pedestrian Crossing Transformer), a multimodal framework that jointly models local and global visual context together with temporal motion dynamics for accurate pedestrian crossing intention prediction. ADAPT processes four spatially aligned visual modalities, including RGB images, local depth maps, global semantic maps, and global depth maps, together with ego-vehicle speed, pedestrian bounding boxes, and skeleton pose information through five specialized modules: a weight-shared Swin Transformer V2 backbone for visual feature extraction, a Cross-Modality Guided Attention module for hierarchical visual fusion, a Mamba-based Motion Feature Encoding module for efficient temporal modeling, a Sparse Cross-Modal Attention module that selectively preserves the most informative inter-modal interactions, and a Vision Transformer-based Temporal Feature Fusion module for sequence-level prediction. Extensive experiments on the JAAD and PIE benchmark datasets demonstrate that ADAPT consistently outperforms existing state-of-the-art methods while maintaining low computational complexity. On JAAD, the proposed method achieves an AUC of 0.73 on JAADbeh and 0.85 on JAADall, while on PIE it achieves an accuracy of 0.92 and an AUC of 0.90. Furthermore, ADAPT performs inference in only 17.23 ms per sample, offering an effective balance between predictive accuracy and real-time deployment efficiency for intelligent transportation and autonomous driving applications. --- ### Rapid-Deployment Crack Measurement Based on SAM3 Semantic-Edge Response Decoding Authors: Zhanping Song, Shipeng Liu, Liang Zhao, Dengfeng Chen Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.12292v4 Reliable crack measurement is essential for infrastructure condition assessment, yet existing image-based approaches typically depend on pixel-wise annotations, task-specific segmentation training, and mask-based geometric measurement, making cross-scene deployment costly and sensitive to segmentation errors. We identify an output-interface mismatch in SAM3: its prompt-conditioned semantic response preserves crack evidence that is often suppressed or spatially distorted in the final candidate masks. Across six public crack datasets, the internal response achieves 82.66% average crack-pixel recall, compared with 74.66% for the retained SAM3 proposals, with an average mismatch ratio of 8.73%. Based on this observation, we propose Semantic-Edge Response Decoding (SERD) to calibrate the semantic response using a fixed Sobel structural field, and further develop SERD-DQ, a training-free framework that directly estimates crack centerline and transverse geometry from the continuous decoded response without generating an intermediate predicted mask. Experiments verify both segmentation fidelity and direct geometric measurement against manually established pixel-level references. Compared with native SAM3 mask-based quantification, SERD-DQ reduces width MAE from 6.072 to 5.547 pixels, length relative error from 22.245% to 17.355%, and area relative error from 33.921% to 26.228%, while achieving a latent geometry recovery rate of 0.394. The results indicate that continuous semantic-edge responses provide a more reliable interface for training-free crack quantification than conventional mask-mediated measurement. --- ### Training-Free Decoding of SAM3 Semantic Responses for Cross-Domain Infrastructure Crack Segmentation Authors: Zhanping Song, Shipeng Liu, Liang Zhao, Dengfeng Chen Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.12292v3 Cross-domain infrastructure crack segmentation is challenged by variations in materials, imaging conditions, crack morphology, and background interference. Although text-promptable foundation models reduce the need for task-specific training, SAM3's native proposal interface may not fully expose the crack evidence already encoded by the frozen model. We identify and quantify an output-interface mismatch in which the internal language-conditioned response preserves ground-truth-aligned crack evidence that is absent from the native proposal output. Based on this finding, we propose \textbf{S}emantic-\textbf{E}dge \textbf{R}esponse \textbf{D}ecoding (\textbf{SERD}), a training-free method that replaces proposal-based readout with direct response decoding. SERD normalizes the dense response, calibrates it using a fixed Sobel prior, and applies a single threshold to produce the crack mask. SAM3 remains frozen without any target-domain annotations. Using one source domain solely for threshold selection, SERD achieves an average Crack IoU of 58.00\% on five unseen datasets, compared with 54.33\% for native SAM3. Across six rotated source-domain settings, it obtains a mean target-domain IoU of 60.23\% and a Boundary F1 of 67.18\%, exceeding SAM3 by 3.27 and 2.70 percentage points, respectively. The results show that task-aligned internal-response decoding provides a simple and transferable solution for cross-domain infrastructure crack segmentation. \textit{Code is available at: \href{https://github.com/xauat-liushipeng/SERD}{GitHub}.} --- ### The Sound of Absence: Audio-Language Embedding Models Struggle with Negation Authors: Chun-Yi Kuan, Hung-yi Lee Published: 2026-07-14 Categories: eess.AS, cs.AI, cs.CL, cs.LG, cs.SD arXiv: https://arxiv.org/abs/2607.12290v1 Audio-language embedding models such as CLAP are widely evaluated on matching present sound events, but rarely on negation. We show this affirmation-only evaluation hides a key limitation: these models fail to encode negated sound concepts, mapping affirmative and negated captions to nearly identical representations. To expose this blind spot, we introduce NegEval-Audio, a framework that converts existing datasets into two negation-aware tasks, Retrieval-Neg and Multiple-Choice Negation (MCQ-Neg), to probe whether models distinguish present from absent events. On AudioCaps and Clotho, performance degrades sharply under negation, with negation-type MCQ accuracy falling far below chance, and the failure persists even for a recent multimodal LLM-based embedding model. While a training-free steering method improves MCQ-Neg, it yields marginal gains for Retrieval-Neg. This indicates that affirmation bias is a fundamental flaw in the representation geometry, necessitating explicit negation-aware training objectives. --- ### GeoSEAN: Explainable Country-Level Image Geolocation for ASEAN Regions Authors: Muhamad Syukron, Danish Rafie Ekaputra, Tintrim Dwi Ary Widhianingsih Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.12284v1 Image geolocation aims to infer the geographic origin of an image from visual content alone. However, this task remains challenging in regions where countries share similar urban, roadside, architectural, and environmental characteristics. Many existing geolocation models focus on coordinate level prediction or classification performance while providing limited insight into how visual evidence contributes to location predictions. This study presents an explainable country level image geolocation pipeline for 11 ASEAN countries. First, we collected 4,850 images from GeoGuessr style sources, Google Images, and additional street level imagery. We then evaluated three approaches on this dataset: CLIP zero shot classification, a LightGBM classifier, and an MLP classifier. The MLP achieved the best test performance, attaining an accuracy and F1 score of 85.91%. For explainability, predictions generated by the MLP classifier were analyzed post hoc using CLIP attention rollout, YOLO26 object detection on the original images, and Energy Based Pointing Game (EBPG) overlap metrics. Object level analysis indicates that frequently detected objects are not necessarily associated with the highest attention density, suggesting that object frequency and attention based visual evidence capture different aspects of a scene. These results demonstrate that the proposed model can support accurate regional image geolocation while enabling object level inspection of the visual cues underlying its predictions. --- ### SlimPer: Make Personalization Model Slim and Smart Authors: Siqi Wang, Xianjie Chen, Shaofeng Deng, Albert Chen, Romil Shah, Jiawei Huang, Zhaoqin Wang, Zhang Zhang, Yiqun Liu, Meilei Jiang, Anish Dubey, Moyan Mei et al. Published: 2026-07-14 Categories: cs.IR, cs.LG arXiv: https://arxiv.org/abs/2607.12281v1 Transformer-style architectures are increasingly adopted for industrial recommendation systems, yet they inherit a design premise misaligned with the task: generative models rely on per-token autoregressive prediction, which justifies maintaining large intermediate tensors that scale with sequence length. In contrast, recommendation systems produce a single set of relevance scores for each <user, item> pair without token-level supervision. Leveraging this observation, we propose SlimPer, which reformulates personalized ranking as iterative refinement of a compact, unified <user, item> knowledge base. At each layer, the model selectively queries raw multi-modal user-side tokens, computes explicit relevance matching scores, and refines the knowledge base, all in O(N) per-layer cost with a fixed-size intermediate representation. As a result, model depth is decoupled from user history length, enabling deeper relevance understanding without proportional growth in compute or memory; request-only optimization further trims memory by sharing a single copy of user-side tokens across all candidate items. SlimPer unifies sparse, dense, and sequence features within a single backbone and provides inherent interpretability through its attention mechanism. Deployed on Instagram Reels and Feed, SlimPer yields measurable improvements in user engagement while streamlining the overall system and enabling effective modeling of 10k+ fine-grained user history events. --- ### A Shared Subcircuit Lets LLMs Count Down Across Tasks Authors: Jacob Dunefsky, Wes Gurnee, Emmanuel Ameisen Published: 2026-07-14 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2607.12279v1 Writing a sentence of exactly twelve words; ending a DNA sequence at the right codon; formatting an ASCII table. These are all tasks that language models can do that requires tracking how many tokens remain before a target. In this work, we identify in Llama-3.1-70B-Instruct a general mechanism for performing these tasks: a "countdown subcircuit" that compares the current position to a goal length and estimates the time remaining until then. We first isolate a countdown subcircuit in a controlled setting, in which the model is tasked with writing a fixed-length sentence ending in a specified word. We then investigate the geometry of the representations used by the subcircuit, and find that the subcircuit uses an identical motif previously identified in a frontier LLM on a separate task, thus suggesting that this motif is shared across models. Finally, we use unsupervised probing on a natural language dataset to find a variety of other tasks where this subcircuit is used, including tasks where the goal length is inferred from context rather than explicitly stated. Our work suggests that reverse-engineering subcircuits allows us to understand how behaviors generalize from a single example to many different tasks and even models. --- ### Auditing Data Leakage in Whole-Slide Image Multimodal Benchmarks Authors: Wenhao Zhang, Zhongliang Zhou, John Kang, Sheng Li Published: 2026-07-14 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2607.12278v1 Recent vision-language models (VLMs) for computational pathology report striking zero-shot performance on whole-slide image (WSI) visual question answering (VQA) benchmarks. We audit these claims and find them fundamentally compromised by data leakage at two hierarchical levels: patient-level leakage, where slides from the same case appear in both training and test folds, and institutional-level leakage, where different cases nonetheless share staining-batch and scanner signatures through a common Tissue Source Site (TSS). By tracing canonical slide, case, and TSS identifiers across major public resources, we document case level train test overlaps of 92.3~100% on TCGA-derived benchmarks, together with near-complete TSS overlap. We further demonstrate that both leakage levels are linearly decodable from foundation-model feature space, that they induce a measurable accuracy gap between leaked and audit-clean cases on a published checkpoint, and that across multiple published WSI VLMs, peak reported accuracies concentrate on the most heavily contaminated benchmarks. Therefore, the current WSI VQA evaluation cannot distinguish genuine multimodal reasoning from nearest-neighbor retrieval over memorized institutional and patient-specific artifacts. Finally, we outline concrete recommendations for contamination-free evaluation. By addressing benchmark construction, provenance disclosure, and automated overlap auditing, we aim to guide future research toward verifiable claims of progress. --- ### Code-MUE: Measuring Code LLMs' Uncertainty through Execution-based Semantic Interaction Graphs Authors: Xiaoning Ren, Yinxing Xue, Lei Ma, Yuheng Huang Published: 2026-07-14 Categories: cs.SE, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2607.12273v2 As Code Large Language Models (LLMs) become central to modern software engineering, their inherent stochasticity poses significant real-world risks, where even minor errors can lead to severe functional, security, or safety consequences. Reliable automation, therefore, demands the ability to distinguish between confident, well-supported predictions and stochastic guessing. However, existing uncertainty estimation methods face a critical gap: white and grey-box techniques are often inapplicable to closed-source models, while standard "black-box" text metrics fail to capture the unique fragility of code, where syntactic variation does not always imply semantic divergence. To bridge this syntax-semantics gap, we introduce Code-MUE, a purely black-box framework that measures uncertainty through execution-based Semantic Interaction Graphs. Different from prior approaches that rely on superficial textual similarity, Code-MUE grounds uncertainty in observable runtime behavior, calculating the Von Neumann entropy of the solution space to quantify global semantic diversity. A large-scale empirical study across eight state-of-the-art LLMs demonstrates that Code-MUE achieves a strong negative correlation with functional correctness (Spearman's correlation up to -0.98), significantly outperforming lexical and embedding-based baselines while enabling robust risk detection and selective prediction in practical workflows. --- ### A hybrid analytical-PINN model for subsurface simulation of geothermal heat exchangers in heterogeneous underground Authors: Moke Rao, Thomas Hamacher, Smajil Halilovic Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.12271v2 Accurate and efficient prediction of subsurface temperature fields is essential for the design and operation of borehole heat exchanger (BHE) systems. Here we develop a parametric hybrid analytical and physics-informed neural network (PINN) framework for long-term multi-BHE simulations in heterogeneous underground. The method analytically extracts the singular line source response and enables the effective training of neural correction associated with subsurface heterogeneity. An explicit parametrization of the thermal conductivity allows physics-informed learning of a single feedforward neural network to generalize across different subsurface conditions. By formulating the correction in borehole-centered relative coordinates, the learned correction can be reused as a universal corrector through spatial and temporal superposition principles. Numerical experiments based on the infinite line source (ILS), finite line source (FLS) and moving finite line source (MFLS) models show that the hybrid method outperforms analytical approximations with stable accuracy over long simulation horizons and achieves orders-of-magnitude speedups over traditional solvers. The proposed framework therefore combines the efficiency of analytical models with the ability of numerical methods to capture heterogeneous subsurface physics, providing a fast and accurate approach for repeated long-term simulation of multi-BHE systems. --- ### A hybrid analytical-PINN model for subsurface simulation of geothermal heat exchangers in heterogeneous underground Authors: Moke Rao, Thomas Hamacher, Smajil Halilovic Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.12271v1 In this paper, a parametric physics-informed neural network for solving the heterogeneous soil thermal problem with borehole heat exchangers (BHEs) as singular sources is developed. There are three novel features in the present framework; namely, (i) the singularity is naturally removed by using analytical line source models; (ii) using the explicit formulation for gradient thermal conductivity enables physics-informed learning of the parametrization featuring the conductivity; (iii) the learned correction is utilized as an efficient universal corrector via superposition principles. We first introduce the decomposition of the temperature change and transform the approximation of the entire heterogeneous response to the correction compensating the difference between the practical solution and idealized homogeneous approximation. In such a way, the delta function singularity is excluded and the bulk heat transfer is captured for the sake of facilitating the effective training of the neural network. The original problem is then reformulated as a governing correction diffusion or advection-diffusion equation subject to a homogeneous initial condition. The linearly varying thermal conductivity is used to model the soil heterogeneity. We propose a physics-informed neural network to approximate a universal corrector with respect to a single borehole with unit heat extraction rate. As a result, the network is trained by minimizing the physics-informed and data-anchored loss function that is evaluated for sampled conductivity parameters on adaptively selected training points. In addition, we include the location indicator function regarding the source as a feature input of network and find that it helps the network to process the local information. We perform numerical tests to exhibit the effectiveness of the proposed method based on three different analytical models. --- ### Quantum Port-Hamiltonian Neural Networks: Learning Conservative and Dissipative Dynamics via Measurement-Induced Nonlinearity Authors: Dibakar Sigdel Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.12269v3 We introduce Quantum Port-Hamiltonian Neural Networks (Q-pHNNs), parameterised quantum circuits that learn classical dynamics in a structure-preserving manner. The framework rests on the Isomorphic Hamiltonian Mapping (IHM): the skew-symmetric interconnection matrix $\mathbf{J}$ corresponds to unitary gate evolution, and the positive-semidefinite dissipation matrix $\mathbf{R}$ to Measurement-Induced NonLinearity (MINL), realised by mid-circuit measurement with classical feedforward. Conservation and passivity are then enforced by construction rather than by penalty terms, and dissipation becomes an intrinsically quantum effect: energy leaves through the act of measurement. We instantiate the IHM in three architectures: a Quantum HNN that extracts Hamilton's equations via the Parameter-Shift Rule; a Q-pHNN that dissipates through MINL; and a topology-entangled Quantum Graph Neural Network lifting both channels to $N$-node coupled-phasor networks. In simulation, where every model here was trained, we obtain $1.35\%$ relative energy drift under a symplectic integrator, $100\%$ energy monotonicity for the single-oscillator MINL circuit, and $92$--$98\%$ phase-space energy decay across ring, star and chain networks at $N\in\{3,6,9\}$. On an IBM Heron processor the trained energy surface and its parameter-shift gradients reproduce their simulated values, with an error budget dominated by readout rather than gate infidelity; the dissipative channel executes natively, but its decay is not separable from measurement back-action at these depths. --- ### Quantum Port-Hamiltonian Neural Networks: Learning Conservative and Dissipative Dynamics via Measurement-Induced Nonlinearity Authors: Dibakar Sigdel Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.12269v2 We introduce Quantum Port-Hamiltonian Neural Networks (Q-pHNNs), a family of parameterised quantum circuits that learn classical dynamics in a structure-preserving manner. The framework relies on the Isomorphic Hamiltonian Mapping (IHM): the skew-symmetric interconnection matrix $\mathbf{J}$ corresponds to unitary gate evolution, and the positive-semidefinite dissipation matrix $\mathbf{R}$ corresponds to Measurement-Induced NonLinearity (MINL) realised via mid-circuit measurement and classical feedforward. This ensures conservation and passivity are enforced by construction rather than penalty terms, and it makes dissipation an intrinsically quantum effect: energy leaves the system through the act of measurement, not through any non-unitary term in the Hamiltonian. We instantiate the IHM in three architectures: (1) a Quantum HNN that learns conservative energy manifolds and extracts Hamilton's equations exactly via the Parameter-Shift Rule; (2) a Q-pHNN that realises dissipation through MINL-mid-circuit measurement of a bath ancilla with conditional feed-forward; and (3) a topology-entangled Quantum Graph Neural Network that lifts both channels to $N$-node coupled-phasor networks, with one bath ancilla per node. Experiments on the nonlinear pendulum and damped harmonic oscillator, and a network scaling study on GPU, demonstrate: (i)~$1.35\%$ relative energy drift with a symplectic integrator and scale correction; (ii)~$100\%$ energy monotonicity for the single-oscillator MINL circuit; and (iii)~$92$--$98\%$ phase-space energy decay from measurement-induced dissipation -- monotone at every step -- across ring, star, and chain networks at sizes $N\in\{3,6,9\}$, alongside exact machine-precision energy conservation in the conservative mode. --- ### Track, Rank, Crack: Epistemic Working Memory Scales Multi-Hop Reasoning in Language Agents Authors: Ning Liu Published: 2026-07-14 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2607.12267v1 Language agents that interleave reasoning and tool use degrade sharply as reasoning chains lengthen, even when each individual step is easy. We trace this to context dilution: an agent's investigative state (what it has confirmed, what it suspects, and what it still needs) lives only implicitly in a growing context window, where early discoveries are buried under later retrievals. We introduce SLEUTH, which makes this state explicit and actionable through a structured epistemic working memory: the agent maintains Confirmed Facts grounded to sources, Active Hypotheses ranked by evidence, and Open Questions that directly drive its next action. Across five multi-hop benchmarks and five established baselines, SLEUTH's advantage grows with difficulty, from +5 points on HotpotQA to +11 on 4-hop chains, surpassing Reflexion without multiple episodes. Analyzing where the remaining gap lies, we identify the evidence sufficiency problem: agents often find the answer but fail to commit, exhausting their budget on needless verification. A lightweight commitment trigger fixes this, but only when the agent already maintains structured state: the identical trigger applied to an unstructured agent yields no improvement, isolating organized epistemic state as the necessary condition for effective commitment. Finally, enforcing protocol adherence on a weaker model recovers up to +19 points on the hardest problems, showing that how an agent organizes its reasoning, not raw model capability, is the active ingredient for scaling multi-hop reasoning. --- ### Saturation Makes Quantization Error Additive: A Coverage Model with a Certificate Authors: Joshua Hill Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.12266v1 Mixed-precision quantization must decide which parts of a model to keep at higher precision. A common premise, shared by sensitivity-based methods such as HAWQ and CoopQ, is that the loss from quantizing a set of layers can be reconstructed from per-layer or pairwise sensitivities measured in isolation. We test this premise at the 4-bit weight-and-activation precisions now being deployed, treating the change in loss $f(S)$ from quantizing a layer set $S$ as a set function on the Boolean cube and analyzing it through two classical changes of basis. This analysis yields two findings. First, across configurations drawn from the deployment distribution, 85--93\% of the variance of $f$ is explained by per-layer effects alone. Second, a monotone transform of a sum of per-layer terms reproduces $f$'s ranking of configurations, misordering at most 2\% of pairs. We propose the coverage model $f(S)=c\bigl(1-\prod_{i\in S}(1-a_i)\bigr)$, which reproduces the measured variance profile of $f$ to within a few percent from its $L$ fitted break-rates. This structure supports two predictors of a configuration's loss, each with $L+1$ parameters. The additive model is the optimal first-order predictor. By Parseval's identity its mean-squared error equals the variance of $f$ left unexplained by per-layer effects, which we measure on full lattices, estimate out of sample at full-network scale, and report with every result as a certificate of how well any additive model can do. The coverage model itself is the second predictor. As allocators at matched memory, they attain the lowest KL divergence among the compared allocators on models from 30B to 355B parameters. Below four bits, the resulting allocations continue to solve code and reasoning tasks at budgets where allocations from gradient sensitivities no longer produce terminating generations. --- ### From Many to Meaningful: Feature-Guided Zero-Shot Chronic Kidney Disease Screening Using Large Language Models Authors: Muhammad Ashad Kabir, Sirajam Munira Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.12260v1 Early screening of chronic kidney disease (CKD) is essential for preventing irreversible progression; however, many machine learning (ML)-based screening methods remain difficult to deploy in community and resource-limited screening settings due to their reliance on large labeled datasets, resource-intensive pathology tests, or high-dimensional clinical features, and limited robustness to population and distributional shifts. This study examines the feasibility of using large language models (LLMs) for early-stage CKD screening in a zero-shot setting, without dataset-specific training. We propose a feature-guided zero-shot framework that evaluates LLM performance using a selected set of clinically meaningful, readily available community-based features, rather than exhaustive clinical inputs. Feature selection was guided by ML-based analysis to identify a compact, clinically relevant subset of variables. Tabular patient records were subsequently serialized into text using standardized prompt templates to enable zero-shot inference. The zero-shot performance of four LLMs (LLaMA-3, Qwen-3, Mistral, and GPT-4o-mini) was evaluated using both the full feature set and the selected subset. Generalizability was assessed across three heterogeneous CKD datasets spanning three countries. Across models and datasets, the selected feature set yielded consistent and statistically significant improvements in balanced accuracy and probability estimates, achieving performance levels suitable for screening purposes. These findings suggest that LLMs can support clinically meaningful, training-free CKD screening using minimal community-accessible patient features, offering a practical complement to conventional ML methods in real-world screening contexts. --- ### On-Device Deep Research at 4B: Exposure Bounds Faithfulness, Retrieval Bounds Coverage Authors: Vinay Kumar Chaganti Published: 2026-07-14 Categories: cs.AI, cs.CL, cs.IR, cs.LG arXiv: https://arxiv.org/abs/2607.12257v1 On-device research agents search a corpus, read sources, and write a cited brief on a personal laptop. Whether their citations are faithful, and at what cost, is unmeasured for a deployable small model. This study fixes one 4B generator on a 24 GB laptop and asks what makes its citations faithful. It separates two quantities usually reported as one number. Cited claim faithfulness asks whether the cited source supports the claim. Trustworthy coverage asks whether the agent also cites the right sources. The study crosses how much of each source the generator sees, 400 against 1500 characters, with the quality of the sources supplied, gold papers against retrieved papers. Two levers fall out, and they act on different outcomes. Exposure sets faithfulness. More of each source lifts faithfulness from 0.45 to 0.58 on retrieved sources and from 0.37 to 0.58 on gold sources, and the two settings converge, so faithfulness is bound by exposure, not by whether the source is correct. The exposure lift is robust to a second, independent judge; the exact convergence is tight under the primary judge and only approximate under the second. Retrieval sets coverage. Trustworthy coverage stays near 0.22 on retrieved sources at any exposure, because recall is held near 0.40, so exposure cannot fix which sources are cited. The extra exposure costs about 235 output tokens. The practical recipe is to raise per source exposure first, cheaply, and then treat retrieval recall as the only remaining lever. --- ### Understanding Structured Health Data through Interaction-Aware Mixture-of-Experts Authors: Ji Hwan Park, Ying Ding, Tianjin Guo Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.12255v1 We study interaction-aware mixture-of-experts for post-stroke rigidity prediction using multi-level views of structured health records. Despite minimal performance gains, routing attribution reveals systematic importance differences across views, underscoring view construction as key to interpretability. --- ### Self-Aware Recursively Self-Improving Agents for Personal Singularity: A Goal-, Scope-, Tool-, and Benchmark-Driven Multi-Agent Architecture Authors: Chengshuai Yang Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.12254v2 Large language model (LLM) agents can plan, use tools, maintain memory, and execute long-horizon tasks. This paper proposes Self-Aware Recursively Self-Improving (SARSI) agents: governed agents that maintain a persistent self-model of identity, goals, capabilities, limitations, uncertainty, relationships, history, and developmental change, and use that model to guide and evaluate recursive improvement. Self-awareness is defined functionally and does not imply subjective experience or phenomenal consciousness. We pair SARSI agents with personal singularity, a bounded human-AI co-development objective in which an agent ecosystem helps a user approach an expanding, user-defined feasible capability frontier. Each agent has a goal contract, bounded scope, validated tool registry, tool tests, end-to-end benchmarks, owner-controlled autonomy, routing, memory, self-model, and improvement policy. A scope router assigns every accepted task to one accountable primary agent and transfers out-of-scope work through structured handoffs. A user-facing Auto-Index selects interactive, hybrid, autonomous, or scheduled behavior without overriding external permissions. The architecture combines a planner-executor-verifier loop, an evidence-gated improvement loop, an external governance plane, decentralized lineages, an owner-directed agent foundry, and a Personal Singularity OS coordinating working, computational-imaging, work-process-learning, and personal-learning agents. We formalize functional self-awareness, scope, routing, improvement acceptance, bounded goal evolution, tool-first execution, and human capability transfer, and provide safety invariants, benchmark design, and a staged implementation roadmap. This is a position and systems-design paper, not evidence that consciousness, unrestricted recursive self-improvement, or personal singularity has been achieved. --- ### FinResearchBench II: A Deep Research Benchmark with Consensus-Derived Gold Rubrics for Distinguishing Financial Report Quality Authors: Beidi Luan, Rui Sun, Sinuo Wang, Yan Gu, Chao Li, Zhenliang Xiong, Jing Li, Zuo Bai Published: 2026-07-14 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.12252v2 Deep research agents are increasingly used to produce long-form financial reports, yet large-scale evaluation remains bottlenecked by the need for human experts to define and execute high-quality rubrics. We address this problem by proposing a scalable pipeline for generating high-quality rubrics without human experts in the final loop. We build a financial deep research benchmark from 104 real-world user queries and automatically synthesize 14,450 query-specific candidate rubrics from model-generated reports. To justify removing human experts from rubric execution, we compare rubric judgments from three human experts with those from a three-LLM judge panel on a sampled subset, and show that LLM-based evaluation is sufficiently consistent with human evaluation to replace it for large-scale rubric screening, including 98.67\% label-level agreement on jointly unanimous items. We then derive consensus-derived gold rubrics through two filters: a strict consistency filter, which keeps a rubric only if the three LLM judges unanimously agree on every report under the same query, and a distinguishability filter, which keeps a rubric only if it assigns at least one majority-yes and at least one majority-no label across the evaluated systems. This process retains 3,687 consistency-passed rubrics, of which 2,600 remain distinguishable and form the final set of consensus-derived gold rubrics. Using this final rubric set, we obtain clearly differentiated rankings across 10 deep research systems, with item-level pass rates ranging from 58.58\% to 22.23\%. More broadly, because the pipeline removes human-expert execution from rubric generation and evaluation, it is naturally scalable for benchmark evaluation, automatic system comparison, and future studies of evaluation-driven system improvement. --- ### When Directional Accuracy Lies: A Base-Rate-Honest Benchmark for LoRA-Adapted TimesFM on Equity Forecasting Authors: Taizhen Cheung Published: 2026-07-14 Categories: q-fin.ST, cs.LG arXiv: https://arxiv.org/abs/2607.12248v2 Large pretrained time-series models such as TimesFM are attractive for financial forecasting, but raw directional accuracy is a misleading scoreboard in equity markets. An early LoRA adapter in this project appeared to reach roughly 80% directional accuracy; we show this is not evidence of skill. Over a long horizon in a rising market, a trivial "always-up" rule attains comparably high accuracy without using the input at all. To separate genuine skill from this base-rate artifact, we build a reproducible, frozen-data benchmark with expanding walk-forward folds, a stratified held-out-ticker split, honest baselines (zero-shot TimesFM, always-up, random-walk, persistence, AR(1)), and paired significance tests (McNemar, Diebold-Mariano) under Benjamini-Hochberg FDR control. We apply the identical method to two universes -- a tech-heavy NASDAQ-100 and a broad S&P 500 -- reporting excess accuracy over the always-up base rate. Three findings replicate. First, when the historical ~80% condition is recreated, the high number is a base rate of ~0.70 that the fine-tuned model scores below. Second, pooled LoRA shows no directional skill over the base rate at any horizon on either universe (negative at the six-month horizon). Third, per-sector specialization is significantly worse than a single pooled adapter (Diebold-Mariano p<0.001 on held-out stocks at h=128). Fine-tuning's only measurable benefit is a statistically significant reduction in point-forecast error relative to zero-shot TimesFM, which nonetheless does not beat naive baselines and confers no tradeable directional edge. The contribution is methodological: a defensible, fully seeded protocol that prevents the base-rate trap, together with the replicated negative result it produces. --- ### Proximity Features: Privacy-Compliant Cold-Start Personalization at Airbnb Authors: Wei Jiang, Bin Xu, Hui Gao, Bharathi Thangamani, Weiwei Guo, Sundar Srinivasavaradhan, Tracy Yu, Huiji Gao, Michael Kinoti Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.12246v1 Personalization in two-sided marketplaces relies heavily on user-level features, yet for platforms with infrequent, high-consideration purchases, a large fraction of users lack sufficient history for effective recommendation, spanning both paid and organic channels. At Airbnb, a substantial share of search requests comes from logged-out or first-time users, with this challenge especially pronounced on paid-channel landing pages, leaving traditional user-level features unavailable for a large fraction of traffic. Privacy regulations and increasing restrictions on third-party cookies further limit identifier-based tracking for non-essential use cases. This paper introduces Proximity Features, a privacy-compliant feature system that groups users by geographic proximity using geo-IP data and an adaptive clustering algorithm, producing aggregated user-level signals for groups of approximately 1,000 nearby users without requiring a persistent individual identifier at inference time. Privacy is preserved by design: the pipeline operates on consented, aggregated data only within consent-gated privacy controls. The system is deployed in production at Airbnb, serving multiple surfaces including marketing landing pages and destination recommendation, with engagement emails integration under way. Online A/B experiments demonstrate statistically significant lifts in bookings, with the largest gains observed among users with absent or stale history. --- ### Rough Path Signature-Guided Geometry Augmentation for Few-Shot Industrial Surface Defect Detection Authors: Jiaqi Kuang Published: 2026-07-14 Categories: cs.CV, cs.LG, math.PR arXiv: https://arxiv.org/abs/2607.12245v1 Few-shot industrial defect detection remains difficult for standard supervised detectors, which achieve poor performance on boundary-dominated industrial defects. This paper proposes rough path signature-guided geometry augmentation (RPS-GA), a geometry-aware approach in which Canny edge contours are treated as ordered planar paths whose truncated second-order signature responses, especially the antisymmetric Lévy-area term, are aggregated into a spatial map that highlights boundary-related structure through two fusion operators, SIG-AUG and SGAA. The approach is evaluated on NEU-DET and PCB-Defect under a few-shot protocol with 5, 10, 20, or 50 labeled images per class, using an unmodified YOLOv8n detector throughout. Compared with the baseline, RPS-GA delivers large gains when supervision is limited, although the margin shrinks as more labels become available. On NEU-DET, SIG-AUG raises 10-shot mAP@0.5 from 0.341 to 0.583, whereas on PCB-Defect, SGAA improves 10-shot mAP@0.5 from 0.086 to 0.299 and yields usable detection at 5-shot where the baseline fails entirely. These trends are confirmed by multi-seed evaluation across independent random partitions. Overall, the results indicate that second-order path-signature geometry offers a practical way to strengthen few-shot industrial defect detection without meta-learning or detector redesign. ---