Detection
-
Foundation-Model-Driven 3D Scene Graphs for Mobile Manipulation
2026-10-16 15:32:09Dominic Maggio, Lorenzo Shaikewitz, Aryan Naveen, Kris Hauser, Luca CarloneAbstract
Geometric and Semantic Foundation models are becoming increasingly popular tools for generating spatial maps for robots. Recent work (e.g., FOUND-IT) has leveraged these tools to construct real-time, task-driven, and open-set 3D scene graphs from a single uncalibrated camera. However, while classical robot mapping techniques have undergone extensive system demonstrations, limited work has been done to develop both the fundamental and the systems wide advancements needed to integrate these new types of robot spatial memory into a full, useable autonomy pipeline. In this paper, we bridge this gap, and in doing so we also evaluate the effectiveness and limitations of this type of 3D memory. Towards this end, we integrate FOUND-IT into a mobile manipulator platform and demonstrate it is well-suited to support multiple objectives such as object change detection, support for multiple mapping sessions, determination of navigable paths, and directly support object grasping for the task of scene re-arrangement. Furthermore, we provide nuanced insights into the effectiveness and limitations of a monocular, foundation-model-driven 3D scene graph construction system using observations from rigorous on-robot experiments. Code is available at https://github. com/MIT-SPARK/FOUND-IT.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23622
PDF
-
Stereo LWIR-RGB Automatic Labeling for Low-Light Camera-Based VRU Detection
2026-10-16 15:31:37Dimitrios Bouzoulas, Pyry J\"appinen, Pejman Habibiroudkenar, Eerik Alamikkotervo, Risto OjalaAbstract
VRU detection in low-light conditions is a key component in ensuring VRU safety during the most collision prone time of the day. Currently, advances in low-light detection are limited due to the lack of large scale datasets and automated labeling methods. Most autolabeling methods use RGB detections as their labels source, which cannot be replicated in low-light conditions. To fill this gap in this paper we present an automated pipeline for low-light VRU labeling. We use LWIR-RGB image pairs and source our labels from LWIR as VRUs differentiate from the background due to their body temperature. Unlike our previous work that relied on a complex hardware setup to align the two modalities, we perform this experiment using a LWIR-RGB stereo setup, lowering hardware complexity and cost, making the pipeline easier to reproduce. We evaluate method end-to-end on a dataset collected in the capital region of Finland. Models trained on our generated labels beat both off-the-shelf baselines and models fine tuned on a hand labeled subset on F1, mAP@50 and LAMR metrics when tested on a manually labeled dataset, and the result holds for two different LWIR detectors. This shows that low-light VRU autolabeling can be made both accessible and effective without specialized alignment hardware.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23621
PDF
-
Not Every Mention Is a Claim: Rethinking Object Hallucination Detection in Vision-Language Models
2026-10-16 15:23:45Hsin-Ling Hsu, Neeraja Kirtane, Kuan-Hao HuangAbstract
Object hallucination in vision-language models (VLMs) is usually measured by checking each object a caption names against the image. Most benchmarks and detectors alike treat every object mention as a claim that the object is present. This assumption fails when a caption names an object without asserting it, as in "a bench for people to sit on", which mentions people without claiming that any are in the image. We call such mentions non-assertive. Treating them as hallucinations inflates the error rates attributed to models that use such language. To separate them from real hallucinations, we introduce NAME-Bench (Non-Assertive Mention Evaluation Benchmark), 25,497 object mentions from captions that five VLMs generate for COCO images. Each mention is labelled concrete (asserted and present), hallucinated (asserted and absent), or non-assertive, and non-assertive mentions are further divided into four subtypes: depictive, purpose, negation and hedged. All the hallucination detectors that we evaluate flag non-assertive mentions as hallucinations, at rates between 14% and 80%. Adding non-assertive mentions to detector training as negatives largely corrects this on LLaVA-1.5. On the four newer VLMs, a comparable correction takes five to ten times more data, and the detectors then catch fewer real hallucinations. These findings show that current detectors conflate mentioning an object with asserting it, and that this conflation is harder to undo on the four newer VLMs than on LLaVA-1.5. Hallucination evaluation should therefore separate what a caption asserts from what it merely mentions.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23610
PDF
-
ADA 2025: A Multimodal Dataset for Water-Deficit Stress Detection in Sugar Beets
2026-10-16 15:23:34Piper Powell, Robert Franz, Amos Smith, Isaak Ihorst, Joachim Hertzberg, Martin AtzmuellerAbstract
Research and development into artificial intelligence (AI) and machine learning (ML) applications in crop management are hindered by a lack of suitable public datasets. Whereas applications in disease and pest detection benefit from a comparatively higher amount of available data, applications like water-deficit stress detection often have very few if any public datasets available. Moreover, most datasets that are available are single-modality and frequently RGB-only, leaving the early detection and enhanced performance offered by other methods, and particularly by multimodal approaches, out of reach. Datasets with accurate metadata, higher numbers of plants, and smaller intervals between data points are also scarce. To address this, we publicly release ADA 2025, a >500 GB comprehensive multimodal dataset of young sugar beet plants under healthy and water-deficit conditions combining thermal, hyperspectral, RGB, and depth imagery along with relevant environmental and metadata and both relative water content (RWC) and leaf thickness as ground truth sources. Exemplary results showcase the dataset's validity in terms of differences detectable between stressed and healthy plants.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23609
PDF
-
Interpreting SSL Representations for Spoof Detection: a WavLM Study
2026-10-16 14:26:55Mohamed Mallat, Michele Panariello, Massimiliano Todisco, Nicholas Evans, Anthony LarcherAbstract
Self-supervised speech models such as WavLM achieve strong spoofing detection performance, yet which layers encode the most spoof-relevant information and what acoustic properties they capture remain opaque. We present a layer-wise interpretability study of WavLM using the ASVspoof 5 database, combining three complementary analyses: (i) per-layer linear probing with attack-level evaluation across speech synthesis, voice conversion, and adversarial attacks; (ii) a canonical correlation analysis (CCA) linking layer representations to spoofing-relevant acoustic features; (iii) an exhaustive grid search over layer subsets showing that use of five layers outperforms full thirteen-layer pooling. CCA analysis reveals that, while most acoustic features peak in alignment with mid-layers and then decay, voice quality measures jitter, shimmer, and the harmonics-to-noise ratio show the opposite trend, with correlation increasing toward later layers. Embedding visualisations reveal that deeper layers capture a continuous acoustic similarity among attacks rather than discrete categories, suggesting these layers encode fundamental synthesis artifacts rather than attack-specific signatures.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23526
PDF
-
FLoTS-SLAM: Optical Flow-Derived Uncertainty for Monocular Dynamic SLAM with Two-Stage Optimization
2026-10-16 14:07:19Haorui Du, Lipu ZhouAbstract
Monocular simultaneous localization and mapping (SLAM) remains challenging in dynamic environments. Real-time dynamic object identification in such environments is a critical problem. Recent approaches estimate per-pixel uncertainty from rendering residuals or multiview feature inconsistency to identify dynamic regions. These methods lack direct motion evidence, and their accuracy may degrade when reconstruction is inaccurate or features are ambiguous. In addition, such uncertainty merely downweights pixels in bundle adjustment (BA), so inaccurate uncertainty estimates can still impair camera pose estimation and 3D reconstruction. Accordingly, we present a monocular RGB SLAM system for dynamic scenes. Specifically, a motion uncertainty estimation method is proposed to derive uncertainty from the residual between optical flow and camera motion induced flow, directly reflecting object motion. The estimation varies with frame alignment accuracy, reducing false detections and yielding accurate pixel-level masks of dynamic regions. To make BA more robust, we estimate camera poses from high confidence pixels. These poses are then used to reproject dynamic pixels during joint refinement of poses and depth, reducing the influence of object motion on both pose and depth updates. Moreover, we provide a quantitative evaluation of dynamic masks and analyze how these masks affect the downstream tracking and reconstruction of dynamic objects. Our approach improves the mean dynamic mask IoU by 35\% over a state-of-the-art baseline on DAVIS, and reduces the average ATE by 59\% on Bonn and 19\% on TUM relative to the SLAM backbone used in our system. The source code will be made publicly available.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23496
PDF
-
Monitor-Validity Failures: LLM Safety Evidence Paths at Evolving Framework Boundaries
Abstract
Modern LLM applications rely on runtime monitors that inspect tool calls, traces, streams, detector outputs, and verifier inputs. This creates an interface-reliability problem: a framework path may remain operational while policy-relevant evidence never reaches its decision point. In one representative case, a regex detector returned a positive match, but the framework stranded the result before the block decision. This differs from jailbreaks and detector errors because the evidence is already recognized yet disconnected from enforcement. We study this repair-boundary signature with canary contracts: parent/fixed confirmation units that place a known payload only on the changed evidence surface and record monitor visibility together with the downstream decision for action-boundary cases. Applying the protocol to public framework repairs, we confirm 18 evidence-path failures across 11 LLM or agent safety project groups and four implementation ecosystems. Nine are established-boundary repairs, eight are new-surface policy-coverage gaps, and one is a boundary-only evidence case. Fourteen reach E1 enforcement boundaries: seven native action outcomes, six configured enforcement boundaries, and one synthetic-consumer consequence. The remaining cases affect verifier, safety-prompt, or observability channels. The cases form carrier, adapter, authority, and surface families and show how ordinary application behavior can remain intact while a safety-specific evidence path misses its decision. Exact revisions, deterministic outcomes, negative controls, and a frozen prospective audit document the confirmation rule. Monitor assumptions therefore need regression tests at the schemas, streams, tools, and adapters that frameworks evolve.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23362
PDF
-
Verbalising Model's Computation for Chain-of-Thought Faithfulness Detection
2026-10-16 10:44:23Yujing Wang, Yulan He, Wenya Wang, Hanqi YanAbstract
Current Chain-of-Thought (CoT) verification methods detect reasoning faithfulness from model outputs or static activations, but are limited in explaining how unfaithfulness develops during generation. We introduce VERITAS to verbalise the model's computation: how different reasoning anchors emerge, transfer across layers, and surface in the output. VERITAS adopts the layer-wise Jacobian projection to produce a verbalised account of the model's internal reasoning evolution at each generated CoT token. By training a classifier on the extracted features along the traces, we show that these structured features provide insightful perspectives on how and where reasoning diverges toward unfaithfulness. Experiments on four datasets from FaithCoT-Bench and BonaFide show that VERITAS achieves a significant performance gain over LLM-as-Judge, with a 28.4% gain in balanced accuracy, and an average improvement of 16.0% over activation probing. Further analysis shows VERITAS can be generalised to different datasets and model architectures. Our work shows that verbalising model computation provides a richer and more interpretable lens on CoT faithfulness, moving beyond detection toward a mechanistic understanding of how reasoning fails.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23275
PDF
-
Who Pays for Efficient Inference? Confidence-Based Small-to-Large LLM Routing Across Tasks and Languages
2026-10-16 10:23:41Md Sabbir Hossen, Anichur Rahman, Pabon Shaha, Airin Afroj Aishi, Andrew H. Sung, Md Shohel RanaAbstract
Small-to-large language model routing can reduce inference cost by escalating difficult inputs while retaining easier ones on a smaller model. Its success, however, depends on identifying examples that benefit from escalation, and the reliability of confidence across multilingual tasks remains unclear. We present a systematic evaluation of confidence-based routing across three model families, four languages, and three task types, covering 72 model, dataset, language, and scale configurations and 42,288 predictions. We compare global confidence routing with random routing, equal-quota confidence routing, and an oracle upper bound. Confidence identifies beneficial escalations on the Belebele reading comprehension and SIB-200 topic classification datasets, with mean correction-detection AUCs of 0.612 and 0.705. At 50% escalation, it improves over random routing by 1.40--4.67 percentage points on Belebele and 2.54--5.63 points on SIB-200. In contrast, the mean AUC on the MMLU-ProX-Lite knowledge and reasoning dataset is only 0.510, and routing provides no consistent improvement over random selection because small models often remain highly confident while incorrect. Global confidence routing also creates language escalation gaps up to 30.33 percentage points, whereas equal-quota routing removes this disparity with little aggregate accuracy cost. These results show that confidence routing is task-dependent and can allocate computation unevenly across languages. Code is available at this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2610.23249
PDF
-
What LLMs Do in the Shadows: Reading Context Reliance from the Unembedding Dark Space
2026-10-16 09:21:49Alexandra Kuleshova, Andrey Volodichev, Daria Kotova, Alexey ZaytsevAbstract
Detecting faithfulness hallucinations in large language models (LLMs) is crucial for their safe deployment in high-stakes domains. Recent approaches address this by contrasting contextual against parametric knowledge utilization, yet measuring context reliance typically demands noticeable overhead, such as materializing attention maps, running additional forward passes, or training auxiliary probes. In this work, we show that context reliance can be estimated by projecting hidden states onto the dark space - directions that compose weakly with the unembedding matrix. Beyond its recognized function as an initial-token attention sink, we show that this subspace actively mediates context processing, and that its share of hidden-state norm - termed the Dark Matter Ratio (DMR) - provides a direct score of context utilization. Pairing DMR with a parametric knowledge score yields DarkLens, a lightweight faithfulness detector operating strictly within a single forward pass. Evaluated across three QA and summarization benchmarks on six open-weight LLMs (4B-13B parameters), DarkLens matches or outperforms a wide range of baselines without requiring extra forward passes or probe training.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23182
PDF
-
Knowledge-Guided Distributional Prompt Learning with Visual Hierarchy for Zero-Shot Anomaly Detection
2026-10-16 09:10:18Jiyul Ham, Seongho Yoon, Yonggon Jung, Sangjae Lee, Jun-Geol BaekAbstract
Zero-shot anomaly detection (ZSAD) aims to detect anomalies in previously unseen classes using auxiliary images, with applications such as industrial visual inspection and medical imaging. Recent vision-language-model-based approaches improve ZSAD by aligning textual and visual representations, while distributional prompting further models prompt variability for unseen categories. However, existing methods do not explicitly combine normality-referenced distributional modeling with hierarchy-specific prompt specialization, limiting their ability to model anomalous semantics relative to normality while addressing both image-level detection and fine-grained localization. We propose Hierarchical Knowledge-Guided Distributional Prompt Learning (HiDist), a normality-referenced hierarchical distributional prompt framework for ZSAD. HiDist models each prompt token with a diagonal Gaussian distribution and generates stochastic prompt variations through Monte Carlo sampling. Rather than learning normal and anomalous distributions independently, anomalous prompt distributions are parameterized as deviations from learned normal references, explicitly encoding their dependency. The resulting distributions are specialized into global and local representations for image-level detection and pixel-level localization, respectively, while an adapted global visual representation guides hierarchical alignment. Experiments on seven industrial and medical datasets demonstrate strong cross-dataset performance across anomaly detection and localization tasks. Extensive ablation and representation analyses further support the effectiveness of the proposed distributional modeling and hierarchy-specific representation learning.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23172
PDF
-
From GAN-Based Models to GPT-5: A Taxonomy-Guided Study of Human Performance in Detecting Deepfake Face Images
2026-10-16 08:41:12Zhiqing Guo, Lixin Jia, Luning Sun, Xiaowei Chen, Guofeng Yu, Liejun Wang, Gaobo Yang, David Stillwell, Thorsten Holz, Shujun LiAbstract
Generative models for deepfakes have rapidly evolved from GAN-based architectures to diffusion models and multimodal large language models (MLLMs). However, empirical research on human performance in detecting deepfake face images has not kept pace. It remains unclear if (and how) detection performance varies across generative models and whether model-specific training improves detection more effectively than general-purpose training. In this paper, we propose the \emph{Deepfake Face Artifact Taxonomy} (DFAT), a cross-model taxonomy of human-recognizable visual artifacts in deepfake face images, based on which we build both general-purpose and model-specific training materials. Through an online experiment ($n = 524$), we examined human performance in detecting deepfake face images and evaluated training efficacy across five representative generative models: three GAN-based, one diffusion-based, and one MLLM (GPT-5). Our results show that participants' detection accuracy was strongly model-dependent and that images generated by more recent and advanced models were not necessarily harder to detect. After training, participants achieved significant performance gains across most conditions, and their improvement varied by model. Notably, model-specific training generally yielded greater numerical gains than general-purpose training. In a five-month follow-up test ($n = 342$), we observed no significant decline in detection accuracy and no significant increase in perceived stress. Overall, our findings provide actionable insights for designing effective, sustainable, and targeted training programs to counter the growing threat posed by deepfakes.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23144
PDF
-
Reliability-Consistent Appearance Modeling for RGB-Infrared Multi-Object Tracking
2026-10-16 08:30:48Bingyang Wang, Jiazuo Yu, Lu Zhang, Lihe Zhang, Huchuan Lu, You HeAbstract
Recent RGB-infrared multi-object tracking (MOT) methods have substantially improved cross-modal representation learning and multimodal information integration. However, most existing methods compress RGB and infrared evidence into a unified descriptor. When one modality is temporarily degraded, the tracker can no longer regulate its contribution separately; unreliable evidence may distort association and contaminate appearance memory. We therefore propose RCA-MOT, a reliability-consistent appearance modeling framework that preserves cross-modal complementarity while retaining post-representation evidence controllability. RGB-specific, infrared-specific, and fused identity representations are maintained as jointly accessible appearance states. For each detection, a simple training-free target-scene estimator produces branch-wise appearance reliability used across the two temporal roles of appearance: relative reliability regulates different identity branches, absolute reliability calibrates borderline candidates near the ReID acceptance boundary, and the same branch-wise reliability controls how strongly matched descriptors update historical appearance memory. The individual components are intentionally simple, focusing evaluation on reliability-controllable appearance organization rather than elaborate fusion or learned reliability-estimation modules. RCA-MOT sets new SOTA results on VT-MOT and UniRTL, improving the previous best HOTA by 5.402/6.075 points and IDF1 by 11.549/12.789 points, respectively, while post-hoc analyses further validate reliability as an informative control signal. Code will be released upon acceptance.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23125
PDF
-
X-MDD: LLM-Based Mispronunciation Detection and Diagnosis for Unseen Languages with Canonical References
2026-10-16 07:50:00Xintong Wang, Junchuan Zhao, Ye WangAbstract
Mispronunciation detection and diagnosis (MDD) provides phone-level feedback to language learners, but extending it to new languages typically requires collecting and annotating learner speech for each target language. This repeated annotation effort is costly and limits the language coverage of pronunciation feedback. We study whether MDD capabilities learned from annotated learner speech can be applied to languages unseen during MDD training when canonical pronunciation references are available. Our multilingual LLM-based model, X-MDD, uses canonical speech and phones as references to predict the phone sequence actually produced by a learner. It incorporates in-context demonstrations for inference-time adaptation and is further optimized with group relative policy optimization (GRPO) using detection F1 and phone error rate. We further analyze how canonical inputs affect prediction and find that (1) canonical speech and phones jointly contribute to actual-phone prediction, and (2) conflicting speech evidence reduces the tendency to copy canonical phones. Trained only on Mandarin and English, X-MDD achieves a detection F1 of 38.07 on Arabic Qur'anic recitation, compared with 44.14 for a supervised challenge baseline trained on Arabic. It also achieves F1 scores of 78.68 and 58.78 on Mandarin and English, respectively, demonstrating effective multilingual MDD within a single model.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23079
PDF
-
SpikeQ: Spike-Guided Post-Training Quantization for Spiking Neural Networks
2026-10-16 07:24:52Kai Sun, Peibo Duan, Zhanglu Yan, Guowei Zhang, Yongsheng Huang, Levin KuhlmannAbstract
On neuromorphic chips, spiking neural networks (SNNs) consume far less energy than artificial neural networks (ANNs), but the limited memory of these chips requires low-bit synaptic weights and membrane potentials. Existing post-training quantization (PTQ) methods provide them without retraining and, like those for ANNs, minimize numerical errors such as the mean squared error (MSE). In an SNN, however, a quantization error reaches later layers only through the spikes it flips. We find that errors of equal MSE can cause very different accuracy losses, whereas errors that flip the same fraction of spikes cause the same loss. We therefore propose SpikeQ, a spike-guided PTQ framework. For weights, we select the level of each weight by simulating the downstream neurons and minimizing their spike errors. For membrane potentials, most existing methods quantize the state after the reset, whereas neuromorphic chips compare the threshold with the membrane potential stored at a fixed bit width. We target this pre-fire quantization and build a reset-aware grid that places the threshold on a rounding boundary, so rounding never flips the spike at the current time step. We evaluate SpikeQ on seven benchmarks of image classification, semantic segmentation, and object detection. With 2-bit pre-fire membrane potentials, every existing grid collapses, whereas SpikeQ retains most of the full-precision accuracy. With both at 4 bits, the models stay within 2.59 points of full precision, shrink their memory footprint 7.8--7.9 times, and use 6.4--6.9 times less energy.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23046
PDF
-
A100: A Hybrid Vision Foundation Model for Agricultural Plant Image Analysis
2026-10-16 06:15:49Mayur Pratim Das, Ajai Dubey, Aksshay Mathew P, Sana Afreen, Rishi Kumar Soni, Deepak Meena, Pratibha Kumari, Mukesh SainiAbstract
The use of accurate vision models for crop monitoring, disease diagnosis, and plant phenotyping in agriculture is hindered by the lack of expertly annotated data, especially in developing countries. We propose A100, an agricultural vision foundation model that is pre-trained using self-supervised learning on 3 million unlabeled images from fields covering more than 100 crop species. A100 uses a hybrid CNN-Transformer backbone that fuses MBConv stages for local feature extraction with Transformer stages for global contextual representation. A100 uses pre-training with the integration of DINO self-distillation, iBOT masked patch reconstruction, and KoLeo uniformity regularization, resulting in features that are better than ImageNet-12K supervised representations for agricultural applications without any fine-tuning. By utilizing a frozen backbone, the accuracy for crop classification of A100 is 95.37%, which is +2.56% better than ImageNet-12K, and reduces the leaf count error by 10.4% on the GroMo25 phenotyping benchmark, proving that the same backbone can be used for classification and regression tasks. After full fine-tuning, the performance of A100 is 99.90% on 32-class crop identification and 99.02% on 44-class disease detection. On the difficult PlantDoc dataset, A100 outperforms all baselines, such as CvT-13 (64.89% against 63.56%). Remarkably, A100 obtains 98.01% accuracy on plant classification using just 1% of the labelled data (about 284 images per category), showing great label efficiency for use cases with limited labeling. Under an identical protocol, A100 also outperforms the publicly released agricultural foundation models FoMo4Wheat and LeafVision by more than 14 points at 1% labels. We plan to publicly release our pretrained model to allow agricultural professionals to adapt A100 for various plant analysis applications using limited annotations.
Abstract (translated)
URL
https://arxiv.org/abs/2610.22991
PDF
-
NL2SQL with Iterative Self-Refinement
2026-10-16 05:39:18Anjali Vemula, Akshat DasulaAbstract
Bridging the gap between human intent and machine logic remains a critical challenge in data access for non-technical users, yet Large Language Models (LLMs) remain unreliable: they hallucinate schema elements, generate syntactically invalid queries, or fail to capture user intent. These limitations are especially critical in domains such as insurance, where KPIs, regulatory constraints, and legacy relational systems demand high precision. We introduce a multi-agent, persona-driven NL2SQL framework that combines (1) domain relevance detection, (2) schema-grounded knowledge construction, (3) KPI-aware SQL synthesis, (4) syntax-level self-repair using sqlite execution feedback, and (5) semantic consistency checking via SQL-to-NL reverse engineering. Our system significantly reduces query hallucination and improves semantic fidelity compared to direct LLM prompting. Evaluated on 60 queries (50 from a HuggingFace text-to-SQL dataset and 10 custom domain queries), the system achieves 90% correctness using Mistral across seven coordinated LLM calls showing that this approach can substantially increase the reliability of NL2SQL pipelines in high-stakes environments.
Abstract (translated)
URL
https://arxiv.org/abs/2610.22968
PDF
-
Beyond Surface Cues: Robust LLM-Generated Korean Text Detection
2026-10-16 05:07:15SungHo Kim, SangKeun LeeAbstract
As large language models (LLMs) generate increasingly fluent text, the need for reliable detection of LLM-generated content has grown substantially. However, existing research remains heavily English-centric, leaving Korean largely underexplored. Moreover, current evaluations of Korean text detection often rely on superficial cues, leading to overly optimistic estimates of detection performance. To address these limitations, we introduce \textsc{KoLing}, a bias-controlled adversarial benchmark comprising 50.9k instances that suppresses superficial shortcuts while preserving semantic content and separately evaluates LLM-generated and LLM-modified text detection. We further propose KoLingProbe, an interpretable detector grounded in theoretical linguistics that models Korean-specific properties across multiple linguistic levels. Experiments show that KoLingProbe substantially outperforms strong baselines on all detection tasks while remaining robust to adversarial attacks. Our analyses further reveal that integrating complementary linguistic features across multiple levels yields both improved performance and interpretability.
Abstract (translated)
URL
https://arxiv.org/abs/2610.22940
PDF
-
DICL-Qwen3VL: Decoupled Confidence Calibration for Interpretable Workwear Detection
2026-10-16 04:46:20Meng Wang, Wu Zeng, Mei Li, Chao HeAbstract
Poor illumination and occlusion obscure clothing cues in indoor work sites, while diverse clothing styles complicate workwear compliance judgments. Although vision-language models (VLMs) support localization and semantic reasoning, their predictions often lack explicit supporting evidence, and a single confidence score cannot distinguish visual uncertainty from ambiguity in compliance judgments. We propose DICL-Qwen3VL, a decoupled interpretable VLM that integrates structured workwear detection, evidence generation, and separate visual and reasoning confidence estimates. First, low-rank adaptation (LoRA) adapts Qwen3-VL to localize visible upper-body clothing and identify workwear. Second, decoupled calibration separates visual observations from compliance reasoning and assigns a confidence score to each. Token entropy and sensitivity to image perturbations provide supervision for visual confidence, linking confidence estimation to the generated observations and their dependence on image content. Finally, group relative policy optimization (GRPO) jointly optimizes detection accuracy, output validity, and confidence calibration. Experiments on the company-collected Indoor Construction Site Workwear Dataset (ICSWD) and the web-collected Indoor Construction Workwear Dataset (WICWD) yield mAP@50 values of 0.730 and 0.753, respectively, outperforming the evaluated LoRA-adapted VLM baselines, including Qwen3-VL. The framework also provides explicit evidence and separate confidence scores to support manual review in industrial safety monitoring. WICWD is publicly available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2610.22929
PDF
-
Bengali Hallucination Detection with Retrieval and Programs: A System Study
2026-10-16 03:46:27Shamim Hasan, Samiya Ahmed ChowdhuryAbstract
Bengali hallucination detection must distinguish unsupported factual claims from incorrect reasoning, often without a supplied reference passage. We present and characterize an inference-only system that combines response-conditioned query rewriting, hybrid retrieval, and task-specific language-model judging. Six retrieval routes search question-answer and passage collections; a router directs each example to a 27B textual judge or a 14B agent that can compare retrieved answers, reason directly, or generate and execute Python. We characterize the system using version-specific code, saved predictions, execution traces, and corpus metadata from the Olikbochon challenge. Kaggle associates a displayed private macro-F1 of 0.956 with Version 2; its saved run processes 2,516 examples in 220.99 minutes on two T4 GPUs. Of 253 math-routed examples, 88 terminate through candidate matching, while execution traces expose 49 stderr-bearing run events across 35 examples. A later run changes six predictions despite identical core worker code. A bounded audit of 1,090 benchmark-related corpus records finds exact question or context overlap for 38 processed evaluation inputs, precluding an unqualified claim of corpus-independent generalization. The contribution is a concrete retrieval-and-computation judging architecture with an auditable operational characterization, not a controlled estimate of component gains. We distinguish the reported leaderboard result from independently recomputed diagnostics and identify the evidence needed for stronger effectiveness and reproducibility claims.
Abstract (translated)
URL
https://arxiv.org/abs/2610.22892
PDF
