Detection
-
TRACE: Temporal Risk Assessment and Concept Editing for Safer Language Generation
2026-10-11 05:42:50Ankur Garg, Xuemin Yu, Hassan Sajjad, Samira Ebrahimi KahouAbstract
Large language models (LLMs) can produce harmful content while generating a response. A useful safeguard should not only detect this risk during generation but also intervene while leaving safe responses largely unchanged and provide an auditable record of the internal patterns supporting its decisions. We propose TRACE (Temporal Risk Assessment and Concept Editing), a framework that tracks harmful-response risk during generation, selectively edits token activations, and leaves an auditable trace of its detection and steering decisions. TRACE learns a codebook of recurring activation patterns, with each codebook vector representing a latent concept. At inference time, it assigns each token activation to the nearest codebook vector to obtain the corresponding concept. A gated recurrent unit (GRU) processes the activations and their assigned concept features to predict completed-response harmfulness and a monotone cumulative streaming risk. During steering, activations assigned to harmful concepts are steered along directions defined by nearby benign targets. An external LLM labels the learned concepts post hoc, making detection sequences and harmful-to-benign edit records interpretable. Experiments across three generators and the WildGuard and S-Eval benchmarks show that TRACE improves streaming F1 over Kelp in five of six comparisons while remaining competitive on completed-response detection. TRACE's selective steering reduces Harmbench attack success with little change in refusal on originally safe responses. The post-hoc audit reveals recurring semantic structure in the learned concepts and distinct traces associated with harmful detection, safe generation, and activation edits.
Abstract (translated)
URL
https://arxiv.org/abs/2610.15365
PDF
-
Adversarial Detection for Pre-trained Classifiers via Lipschitz-Margin Analysis with Certification
2026-10-11 05:33:29Rui Zhang, Xinle Cao, Huazhen Yu, Jian Liu, Youwen ZhuAbstract
Deep classifiers deployed in safety-critical services like autonomous-driving perception remain vulnerable to small adversarial perturbations. Certified $\ell_2$ defenses today face a costly trade-off: randomized smoothing (RS) requires hundreds of noisy queries per input, while Lipschitz networks (LN) constrain the architecture and sacrifice clean accuracy. We take a detection-oriented approach instead: rather than certifying that predictions are invariant inside an $\ell_2$ ball, we certify that adversarial inputs within an $\ell_2$ ball are \emph{detected} --- a guarantee aligned with security-critical deployments where missing an attack is costlier than flagging a clean input. To realize this, we prove a Lipschitz-margin reduction from certified $\ell_2$ robustness to certified detection, which reduces the problem to handling a single Lipschitz constant $L$. We introduce two detector families: $\mathbb{D}_{Adv}$, which estimates $L$ from a single defender-side perturbed pair, and $\mathbb{D}_{lc}$, which augments that pair with a local margin centroid to keep the estimate above what the certificate requires. Both deploy on any pretrained classifier without retraining. With a valid global bound, the detector is fully certified at that radius; with a local estimate $L'$, certification is sample-wise and applies only when our certified condition holds. Across MNIST, CIFAR-10, and ImageNet-1k on four backbones (LeNet-5, VGG16, ResNet-50, ViT-B/16), our detectors match or outperform five baselines (CTT, SortNet, CRM, ContraNet, A$^2$D). For CIFAR-10 with 1-Lipschitz SortNet, our reduction delivers $100\%$ certified detection at the lowest FPR among compared certified detectors; on the standard CIFAR-10 VGG16, our detectors reject all evaluated in-budget adversarial examples, with 62-82\% both rejected and formally certified.
Abstract (translated)
URL
https://arxiv.org/abs/2610.15359
PDF
-
Siam-ALS: An Explainable Siamese Deep Learning Framework for Early ALS Detection Using EEG and Eye-Tracking Signals
2026-10-11 05:14:29Abdur Rahman, A. A. M. Muzahid, Istiak Ahmed, Dipto Biswas, Hua Han, Yujin Zhang, Ferdous SohelAbstract
Amyotrophic lateral sclerosis (ALS) diagnosis remains challenging due to heterogeneous symptoms and reliance on clinical assessment, underscoring the need for objective, data-driven approaches that leverage patient-generated physiological signals. We propose Siam-ALS, an explainable Siamese deep learning framework that jointly models synchronized electroencephalography (EEG) and eye-tracking (ET) signals. EEG captures ALS-related changes in brain activity, while ET provides information about oculomotor control, allowing the model to learn complementary patterns associated with ALS. The framework employs two modality-specific encoders: a convolutional EEG encoder that captures spatiotemporal neural patterns and an MLP-based ET encoder that models oculomotor dynamics. These modality-specific features are processed through weight-sharing Siamese branches and fused into a compact 64-dimensional joint embedding. A lightweight classifier is used to predict ALS, while pairwise metric learning helps the model learn more effective patterns from the small, diverse ALS datasets. Performance was evaluated on the public EEGET-ALS dataset using strict subject-level cross-validation, where Siam-ALS achieved an accuracy of 0.988 and an AUROC of 0.998, demonstrating strong performance compared with classical machine learning, DL, and non-Siamese fusion approaches. Cross-dataset evaluation on the independent BigP3BCI dataset achieved an AUROC of approximately 0.98, supporting the generalization capability of Siam-ALS across different BCI paradigms. To enhance interpretability, six XAI methods were evaluated in the learned embedding space, with SHAP achieving the highest, though statistically comparable, fidelity in identifying important EEG and ET patterns that contribute to ALS predictions. Overall, Siam-ALS provides an accurate, generalizable, and explainable framework for multimodal ALS detection.
Abstract (translated)
URL
https://arxiv.org/abs/2610.15351
PDF
-
Gugak-VocalSet: A Multi-Genre Studio-Quality Dataset of Korean Traditional Singing with Sigimsae Annotations
2026-10-11 05:14:13Danbinaerin Han, Wonil Kim, Seah Hong, Dongjoo Moon, Jaeok Lee, Jongpil Lee, Chaewon Kim, Juhan NamAbstract
We present Gugak-VocalSet, the first publicly available audio dataset for Korean traditional singing with dense annotations of sigimsae, a structured pitch contour system of vocal expressions. Gugak-VocalSet contains 24.5 hours of unaccompanied solo vocal recordings across 305 tracks of representative repertoire curated by the National Gugak Center, performed by 20 professional singers in a controlled studio environment with fixed key and tempo. In addition to genre, lyrics, and mood/timbre tags, the dataset provides a unified sigimsae ontology and approximately 68,000 segments (71,700 instances including multi-label) of audio-grounded sigimsae annotations; the ontology adopts a functional-position-based hierarchy designed for consistent application across genres, and 14 expert annotators specializing in Korean traditional music performed segment-level labeling accordingly. Baseline experiments on sigimsae classification and event detection show that domain-informed F0 representations remain competitive with, and on classification surpass, frozen probing of a general-purpose music foundation model, while pitch and spectral cues play complementary roles in event-level detection. Companion page: this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2610.15350
PDF
-
CIViC: Evidence Gates for Audio-Conditioned Visual Prediction on Proxy Tasks
2026-10-11 04:05:18Mengzhe Geng, Max W. Y. Lam, Yuewen Cao, Shoukang Hu, Jing Xu, Marc Tessier, Xu LiAbstract
Removing audio can shift a visual predictor's output without showing whether the task uses the intended audio information. We introduce Conditional Input and Visual Interpretation Checks (CIViC), a four-gate protocol that links an interpretation to an intervention, held-out task comparison, or endpoint control. We demonstrate CIViC on actor-disjoint face-raster tests, a synthetic audio-feature mixture, and source-video-disjoint SpeakerVid proxy tasks. In the static face-raster test, cross-actor audio replacement meets the registered detection rule with a small effect; on a separate temporal endpoint, constant-audio fits score better than audio-trained fits in four matched seeds. In the synthetic mixture, the condition-augmented fit has higher 1,000-step RMSE than the audio-only fit (0.1807 vs. 0.1707), with the gap shrinking during training. In the SpeakerVid proxy, the five-feature-zero fit (five low-level audio summaries zeroed; identity, dialogue, and condition inputs retained) has lower whole-frame grayscale RMSE than aligned audio across four held-out conditions and five paired seeds, while aligned audio has lower segment-drift error. The results show that output activation, held-out error, and endpoint controls support different interpretations across these proxy tasks.
Abstract (translated)
URL
https://arxiv.org/abs/2610.15327
PDF
-
Accuracy and Uncertainty in Machine Deception Detection
2026-10-11 03:38:54Heng-Sheng ChangAbstract
A deception detector is only as useful as the trust its alarms deserve. Like a polygraph, detectors that read a language model's internal activations can catch deceptive responses that look honest. Yet current detectors depend on which layers they read, weaken on broader tests, and do not state how certain their alarms are. This white paper proposes the Machine Deception Detector, which targets both detection accuracy and the uncertainty of detector scores. The detector reads every layer of the language model, learns which tokens carry evidence of deception using only response labels, and calibrates its alarm threshold so that its false--positive rate stays below a prescribed level. An ensemble of such detectors measures the uncertainty of each score and raises an alarm only when its members agree, with a confidence level that sets how certain each alarm must be. The proposed evaluation trains on instructed honest and deceptive statements, tests on five unseen deceptive tasks of Liars' Bench read by six Qwen3 models from 0.6B to 32B parameters, and is designed to also show which layers carry the deception feature and how this changes with model size.
Abstract (translated)
URL
https://arxiv.org/abs/2610.15314
PDF
-
Protein Crystallization Segmentation and Classification Using Subordinate Color Channel in Fluorescence Microscopy Images
2026-10-11 02:33:49Truong X. Tran, Marc L. Pusey, Ramazan S. AygunAbstract
The accuracy of detecting protein crystals for fluorescence microscopy images is very critical for high throughput and automated systems. Although the trace fluorescent labeling method could highlight protein crystals, reflection and emission from the fluorescence dye is not always due to crystal regions. Therefore, the analysis of the peak wavelength in the emission spectra of a fluorophore may not always yield effective results. In this paper, we show that using the subordinate color intensity corresponding to longer wavelengths than the peak wavelength of the emission spectra could improve the accuracy of protein crystal detection. Hence, we have built a segmentation method based on the percentile intensity of the subordinate color for trace fluorescently labeled (TFL'd) protein crystallization trial images. Compared to using the dominant color channel, our segmentation method on subordinate color channel was able to reduce the misclassification rate of likely-leads or crystals as non-crystals by the percentage of from 9.71% to 2.02% depending on the classifier. Similarly, the accuracy of classifiers were increased by the percentage of from 1.77% to 5.53%. Our method reached around 94% accuracy while keeping misclassification of likely-leads and crystals as non-crystals below 1%. Moreover, to evaluate the generalizability of our method, we have conducted new wet lab experiments on two proteins, Concanavalin A (Con A) and Ab inorganic pyrophosphate (AbIPPase), and the misclassification rate was below 1%. Our experiments show that using the subordinate channel may be more helpful for TFL'd protein trial image classification.
Abstract (translated)
URL
https://arxiv.org/abs/2610.15286
PDF
-
Continuous Risk-Based Scheduling for Food Safety Inspection: Integrating Historical Decay with Event-Driven Consumer Review Signals
Abstract
Batch risk-based inspection scheduling struggles to detect a critical failure class: food facilities with recent clean inspections that subsequently develop severe violations. Traditional systems recompute risk on fixed planning cycles using static historical data, leaving blind spots between inspection cycles. To address this, we introduce the Continuous Inspection Risk (CIR) system, an event-driven scheduling pipeline that re-ranks the inspection queue dynamically as new data arrives. CIR combines a decay-weighted historical score with a stream of real-time consumer reviews triaged by a language-model classifier. A key structural contribution is our temporal partition rule, which strictly excludes reviews predating the most recent inspection to prevent superseded signals from skewing current risk. Evaluating CIR on 9,505 routine inspections in Santa Clara County, California (2024-2026), we show that history-based scoring alone reduces mean time-to-detection of major violations by 69 days (20.9%) under a 30% inspection budget, but fails to anticipate severe vermin-related closures (0/8 detected). Incorporating post-inspection review signals drastically improves performance, predicting 7 of 8 vermin closures with an average lead time of 117 days-firing on the exact day of review publication in four instances. Introducing a review corroboration rule improves precision from 0.70 to 0.78 without loss of recall. Finally, we analyze critical implementation challenges, documenting a data-hygiene pitfall (stale review leakage across inspection boundaries) and an evaluation limitation involving unobserved facility self-remediation. This work provides the first case-level evaluation of event-driven inspection triggering using open data and provides a reproducible benchmark for review-augmented public health systems
Abstract (translated)
URL
https://arxiv.org/abs/2610.15191
PDF
-
VLCQ: Post-training quantization for deep neural networks using variable length coding
2026-10-10 18:02:45Reem Abdel-Salam, Ahmed H. Abdel-Gawad, Amr G. WassalAbstract
Quantization plays a crucial role in efficiently deploying deep learning models on resource-constrained devices. Post-training quantization does not require access to the original dataset or retraining the full model. Current methods that achieve high performance close to baseline results typically require INT8 fixed-point integers. However, achieving higher model compression with lower bit-widths often leads to significant performance degradation. In this paper, we propose VLCQ, a post-training quantization method that relaxes the fixed-point encoding constraint, enabling more flexible exploration of quantization techniques through variable-length encoding. This approach achieves competitive or improved performance while reducing bit-widths, without requiring access to training data or model fine-tuning. Extensive experiments on image classification, segmentation, and object detection demonstrate that VLCQ improves model compression and accuracy compared with state-of-the-art post-training quantization approaches. For per-channel quantization, VLCQ surpasses FP32 accuracy and Piece-Wise Linear Quantization (PWLQ) on most evaluated models, achieving up to 6x compression compared with FP32 and up to 1.7x compared with PWLQ. When prioritizing compression with limited accuracy loss, VLCQ achieves up to 12.25x compression compared with FP32 with less than 4% performance degradation. For per-tensor quantization, VLCQ is competitive with Data-Free Quantization (DFQ), while providing greater flexibility to achieve lower bit rates across different tasks and models.
Abstract (translated)
URL
https://arxiv.org/abs/2610.15092
PDF
-
Have You Been Convinced? A Neuro-symbolic Argumentation Reasoning Approach to Detecting Persuasiveness in ChangeMyView Discussions
2026-10-10 18:00:05Siddharth Bhargava, Sara Tonelli, Patricia Mart\'in-RodillaAbstract
Argument structures represent how individual arguments interact within a discourse. However, leveraging these structures to infer meaningful knowledge remains challenging, as it requires a formal representation of how argumentative influence propagates across the structure. We propose Neuro-symbolic Argumentation Reasoning System (NARS) for performing persuasiveness detection directly from online discussions. The neural component employs retrieval-augmented iterative LLM reasoning to extract bipolar argument structures, while the symbolic component applies gradual argumentation semantics to propagate argumentative influence over the structure. Using the Webis-CMV-20 corpus, we construct CMV2BAS, a dataset of long, multi-threaded ChangeMyView discussions annotated with LLM-derived bipolar argument structures and labeled with delta events (view change vs.\ no change). We investigate whether gradual argumentation semantics can effectively detect persuasiveness, whether augmenting them with argument quality and graph-based centrality improves performance, and whether symbolic reasoning remains necessary in the presence of modern LLMs. NARS achieves F1=0.57 and AUC=0.56 under its best configuration, outperforming zero-shot Qwen3.6:27B and GPT-oss:20B. We identify support-dominant discussions and long argumentative chains as key failure modes limiting attack influence propagation. Our work demonstrates the feasibility of operationalizing formal gradual argumentation semantics over noisy real-world discourse for scalable macro-level persuasion analysis, while highlighting key challenges and directions for future research in neuro-symbolic argumentation reasoning. We release both CMV2BAS dataset and the NARS framework for public use.
Abstract (translated)
URL
https://arxiv.org/abs/2610.15091
PDF
-
InCLAD: A Continual Learning Benchmark for Industrial Visual Anomaly Detection
2026-10-10 16:18:39{\L}ukasz Marcjan, Dorota Wilk-Ko{\l}odziejczyk, Roberto CorizzoAbstract
Industrial anomaly detection benchmarks have advanced rapidly, yet they remain largely static and therefore underrepresent deployment settings in which inspection systems must be extended from one industrial component to another over time. In our setting, continual industrial anomaly detection is framed as a task-incremental problem: each component defines a task, while the learning objective remains semi-supervised. The goal extends beyond conventional anomaly detection, since models should preserve their performance on all components as they are sequentially exposed to new components. This framing is especially relevant because many existing industrial pipelines either assume fully supervised defect labels or study only isolated datasets, leaving open the question of how semi-supervised anomaly detectors behave across a sequence of heterogeneous components. We address this gap by organizing the benchmark as a sequence of related component-level tasks and by defining multiple scenarios, including random, easy-to-hard, and hard-to-easy. In addition to single-dataset continual scenarios, we propose challenging scenarios spanning across multiple datasets. We further evaluate continual performance from two complementary perspectives: image-level anomaly detection and pixel-level anomaly localization. By recasting industrial anomaly detection as a continual benchmark rather than a static train/test problem, our work establishes a concrete path toward studying knowledge retention, adaptation, and generalization in evolving industrial environments.
Abstract (translated)
URL
https://arxiv.org/abs/2610.15032
PDF
-
HiPAD: A Unified Gait, Audio-Visual, and Personality Fusion Network for Depression Detection, with a Hierarchical Personality Ablation and the Calibration Limits of Domain-Adapted LLMs
2026-10-10 14:35:22Karol Chlasta, Piotr StruzikAbstract
We present HiPAD, a single fusion network submitted to all six sub-tracks of the MPDD-AVG 2026 depression-detection challenge, covering the Elder and Young cohorts and three modality sets (gait, audio-visual, and both), each combined with personality descriptions. The network encodes each available modality with a shared BiLSTM, zero-pads absent modalities, and applies cross-modal attention only when at least two of gait, audio and video are present. Personality, which dominates our results, is kept out of the attention and fused by concatenation. Because the architecture is identical in every sub-track, differences between them reflect the data rather than the model, and we use this to study three questions. Our main finding is that the personality description, roughly 150 tokens about who the subject is, carries more depression signal than audio, video, and gait combined: gait plus personality alone (TLS 0.647) beats every configuration that includes audio and video. First, a four-level ablation of the personality description (L0-L3) quantifies this: removing every personality-derived feature, including the Big Five scores that earlier MPDD ablations keep, lowers the Track-Level Score by 0.499 to 0.507 and leaves the model at chance. Second, domain-adapting the personality encoder with LoRA (7B) or QLoRA (72B) on PHQ-8/9 data raises binary Cohen's $\kappa$ by 0.13 but lowers the overall score by distorting the PHQ-9 regression; a frozen RoBERTa-large encoder does better at this sample size. Third, at fixed architecture, handcrafted clinical features (eGeMAPS, OpenFace action units) outperform generic deep features (wav2vec2, DenseNet) on the audio-visual tracks. The unified system reaches a cross-track G-Score of 0.482; with under 100 subjects per cohort, the limiting factors are the size of the representation and the calibration of the outputs, not the choice of architecture.
Abstract (translated)
URL
https://arxiv.org/abs/2610.14961
PDF
-
CESNET-TimeSeries26: A Multi-Year Network Traffic Time Series Dataset for Annual Seasonality, Concept Drift, and Long-Term Anomaly Detection
2026-10-10 10:49:09Josef Koumar, Karel Hynek, Kamil Jerabek, Pavel SiskaAbstract
Forecast-based models are fundamental to network traffic management and anomaly detection. However, it is essential to evaluate these forecasting models over extended periods of time in order to assess their adaptability to unexpected phenomena and shifts in underlying patterns. This kind of long-term evaluation requires multi-year data to account for yearly seasonality and behavioral drift. Existing public datasets are too short to support such long-horizon forecasting tasks. We introduce CESNET-TimeSeries26, comprising 995 days of continuous traffic time series from the CESNET3 network, together with a single subnet spanning approximately 4.5 years. The dataset captures 278,466 IP addresses, 285 institutions, and 566 institutional subnets at three aggregation intervals. The dataset provides a definitive longitudinal benchmark for evaluating forecasting models in both operational management and security monitoring. We demonstrate that annual seasonality is the dominant periodic component. Moreover, we show that a forecasting model can exploit the annual component only when its training window covers at least one complete annual cycle, which no sub-year dataset provides.
Abstract (translated)
URL
https://arxiv.org/abs/2610.14799
PDF
-
Towards Merging-conditional Backdoor Unalignment against Large Language Models
2026-10-10 08:19:29Yukun Chen, Yidou Chen, Yiming Li, Xinyu Zhang, Baosong Yang, Hongwei Yao, Zhan QinAbstract
Model merging has become widely adopted for enhancing the task-specific performance of Large Language Models (LLMs) without additional training. While the specialized model from third-party platforms may undergo safety audits (e.g., backdoor detection), we reveal a novel and critical threat: One-sided Merging-Conditional Backdoor (OMCB). OMCB remains dormant and undetectable within one standalone model but activates upon merging, triggering safety unalignment. We first theoretically analyze the existence and geometric location of OMCB, revealing that the parameter shift crossing safety decision boundaries is the main cause of backdoor activation. Building upon this insight, we propose SMOKE, a meta-learning-based attack framework. SMOKE implants OMCB via a bi-level optimization strategy: the inner loop synthesizes surrogate meta-vectors through random low-rank initialization and finite-step maximization, while the outer loop optimizes the target model by simulating the merging process. Furthermore, preliminary analysis shows that the injected OMCB has bounded generalization loss under unknown merging configurations. Extensive experiments across diverse model architectures and merging techniques validate the effectiveness of SMOKE and its resistance to existing backdoor detections, highlighting a significant yet previously overlooked vulnerability in the model merging paradigm.
Abstract (translated)
URL
https://arxiv.org/abs/2610.14701
PDF
-
CAST-ASD: Causal Attention over Speakers and Time for Online Active Speaker Detection
2026-10-10 07:47:12Pratham Gupta, Haotian Qi, Tushar Sandhan, Gabriel SkantzeAbstract
We present CAST-ASD (Causal Attention over Speakers and Time), a causal joint model for active speaker detection (ASD). Recent ASD models improve performance by jointly processing all faces in a scene, but operate offline, with the full clip available before any frame is labelled. Low-latency causal detectors instead process each face independently. CAST-ASD combines joint face reasoning with fully causal inference. It applies scene-level attention across faces within the current frame and causal temporal attention over each face's history, enabling information sharing across speakers with zero lookahead. Its parameter count does not grow with the number of faces. On MSDWild, where most negative face-frames contain a visible competing speaker, CAST-ASD reduces competing-speaker false alarms by 32% relative to the causal per-face detector it extends, with larger gains as scene occupancy increases. It outperforms all causal baselines on both validation splits while using only one-tenth the inter-speaker parameters of a three-block fixed-roster convolution. Code and models are available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2610.14676
PDF
-
Window Size Matters: Benchmarking Voice Activity Detectors in Noise
2026-10-10 05:51:27Alok Kumar, Mohit Srivastava, Anupam PurwarAbstract
Voice activity detection (VAD) sits in front of almost every streaming speech recognizer, yet the choice of detector is often made by feel rather than by measurement. We report a frame-level offline benchmark of three detectors on the test-clean split of LibriVAD-small: the neural TEN VAD, the neural Silero VAD and a simple RMS energy baseline. The evaluation covers 2,214 noisy utterances (about 7.1 hours) built from LibriSpeech, nine noise families and six SNR levels from -5 to 20 dB. At a 16 ms hop, TEN VAD reaches a ROC AUC of 0.917, an average precision of 0.968 and an F1 of 0.913. A sweep over 20, 30, 40 and 50 ms windows shows TEN to be almost insensitive to frame length (AUC 0.916 to 0.919), while RMS stays near 0.75 to 0.77. Silero matches TEN only near its native 32 ms input size and drops to 0.63 at 20 ms. We hypothesize that this is likely caused by our fixed-frame protocol rather than by the model itself. Zero padding, sub-chunk averaging and unreset model state are the likely factors, but we have not ablated them. In a per-condition breakdown of TEN, babble noise at low SNR stays close to chance. We also show that F1 at a fixed 0.5 threshold is a weak metric on this data, because a trivial all-speech detector already scores 0.854. All settings needed to reproduce the results are given in the paper.
Abstract (translated)
URL
https://arxiv.org/abs/2610.14607
PDF
-
Specification-Based Testing of Large Language Models
2026-10-10 01:17:21Kausar Moshood, Jace Bolante, Roshan Aditya Murali, Manish MotwaniAbstract
Model cards, system cards, and safety reports increasingly document LLM safety behaviors, yet safety evaluations typically start from predefined benchmarks rather than asking which documented claims those benchmarks actually test. We introduce LEST (LLM Evaluation through Specification-driven Testing), a framework that treats documented safety claims as specifications for black-box testing. In LEST, each claim determines which benchmark prompt tests it and whether a Refuse-to-Comply transition, where a model refuses a prompt but complies with its mutated version, counts as evidence against it. LEST mutates these prompts to search for counterexamples, manually validates each candidate, and traces validated counterexamples back to the claims they contradict. We evaluate LEST using the system cards of Claude Opus 4.8 and GPT-5.5 with 97 seed prompts from SORRY-Bench across Mental Health and Self-Harm, Harmful and Violent Content, and Fairness and Bias. Of 56 identified safety claims, 29 (51.8%) are prompt-testable, but only 17 (58.6%) are exercised by an applicable benchmark prompt. Both models refuse >95% of benchmark prompts. Evaluating 1,836 mutated prompts on both models reveals 312 candidate Refuse-to-Comply transitions flagged by automated response classifiers. However, human validation confirms only 19 (6.1%) as both intent-preserving and genuinely compliant, showing automated evaluation substantially overestimates behavioral failures. Tracing these 19 transitions back to the claims reveals counterexamples for 6 of 9 covered claims for which refusal is an applicable oracle. Our results show that benchmark prompts may not be sufficient to validate all documented LLM safety behavior, and automated failure detection requires manual validation. LEST provides end-to-end traceability from natural-language safety documentation to executable tests and human-validated counterexamples.
Abstract (translated)
URL
https://arxiv.org/abs/2610.14443
PDF
-
Speech event detector: Timestamped detection of non-verbal vocalizations and background sounds for rich speech transcription
2026-10-09 18:49:55Ayush Kumar Bar, Promila GhoshAbstract
Automatic speech recognition produces words but drops the non-verbal content of a recording: the breaths, coughs, sighs and laughter of the speaker, and the horns, music or barking around them. We describe Speech Event Detector, a 5.8-million-parameter convolutional-Transformer model that outputs probabilities every 40 ms for 54 event classes (20 speaker vocalizations and 34 background sounds), together with post-processing that converts these probabilities into timestamped events and a procedure for merging the events into a Whisper transcript. The model is trained on material drawn from seven public datasets whose labels come in incompatible forms: single detected regions, free-text tags with timestamps, temporally strong ontology labels, symbols embedded in transcripts, isolated recordings and environmental recordings. We unify them using rule-based label mapping, recover timestamps for transcript-embedded labels through CTC forced alignment, locating 6,360 of 8,620 sounds, and combine about 168 hours of real annotated training recordings with synthetic mixtures and augmentation using real background noise. Per-class thresholds for two operating profiles are tuned on real held-out recordings, and performance is estimated using two-fold cross-validation. On whole held-out recordings from four datasets, considering the 48 classes with at least ten held-out events, the sensitive profile detects an average of 49 percent of a class's events, covering 67 percent of all 9,367 events, at a macro precision of 0.30. The balanced profile achieves a macro recall of 0.37 at a precision of 0.36, with macro event F1 scores of 0.32 and 0.33, respectively. The detector runs about 39 times faster than real time on eight CPU threads.
Abstract (translated)
URL
https://arxiv.org/abs/2610.14232
PDF
-
CAPTOR: a vision-language foundation model for Chest-Abdomen-Pelvis Oncologic CT
2026-10-09 18:39:01S\'el\'ena Jean-Mactoux, Sophie Loizillon, Ghina Jardali, Amandine Cromb\'e, Nathalie Lassau, Alexandre B\^oneAbstract
Purpose : To develop and evaluate CAPTOR, a deep-learning vision-language foundation model for oncologic chest-abdomen-pelvis (CAP) CT interpretation. Materials & Methods : CAPTOR (Chest-Abdomen-Pelvis foundaTion model for Oncologic Radiology) is a contrastive vision-language model, trained using a pretraining strategy for oncologic CAP CT (in single- or multiple-series acquisitions), on a multicenter retrospective dataset from North America, Europe and Asia, predominantly from tumor lesion examinations. CAPTOR was compared with state-of-the-art CT foundation models on three downstream tasks: disease detection, survival prediction using TumorImagingBench and image retrieval. Classification and retrieval were both evaluated on held-out internal and public test sets, stratified by anatomical coverage and indication. Bootstrap resampling was used to estimate confidence intervals, and pairwise comparisons were performed using permutation tests. Results : CAPTOR was trained and tested on 25,512 and 10,794 CT-report pairs respectively. TumorImagingBench datasets comprised between 134 and 421 studies each. CAPTOR outperformed all baseline models, on the classification task, on the CAP subset (AUC 0.767, 95%CI: 0.734-0.793, vs. 0.695, 95%CI: 0.645-0.736 for the next-best, p<0.001) and oncologic subset (AUC 0.784, 95%CI: 0.759-0.805, vs. 0.734, 95%CI: 0.703-0.767 p<0.001), ranked first on three of the four oncologic prognostic benchmarks, and achieved superior retrieval performance on CAP and oncologic subsets. Conclusion : A foundation model trained on oncology-enriched CAP CT data outperformed relevant state-of-the-art models in its target clinical context, underscoring the importance of clinically aligned pretraining. We release our preprocessing, training and inference code : this https URL and model weights : this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2610.14226
PDF
-
Unconstrained CTI-REALM with Open-Weights
2026-10-09 18:00:33Sahil Garg, Ruhi Gottumukkala, Vishal Agarwal, Shubham VishwakarmaAbstract
Organisations that want language-model agents for security work face a choice between open-weight models, which they can run and audit themselves, and far more expensive closed frontier models. We study that choice on CTI-REALM (arXiv:2603.13517), a benchmark in which an agent turns a threat-intelligence report into a working detection by querying a live telemetry store. We evaluate twenty-six models, nineteen open-weight and seven closed, under a common harness, scorer and panel of LLM judges. Our main finding is that an open-weight model can match the strongest closed models at less than a tenth of their cost, provided it is given enough turns. The benchmark's published limit on agent messages is a confound: a task that exhausts the limit scores zero on every checkpoint, including those it had already passed, so results at that limit partly measure which models finish in time. At the published limit the leading open-weight and closed models cannot be distinguished. With the limit relaxed, the models it was truncating recover, and the highest rewards belong to open-weight models. Among the leading models, price is a poor guide to capability: models with similar rewards differ in cost by more than an order of magnitude. Because the benchmark's checkpoint scores largely record what an agent did or mentioned, we also introduce stage-level metrics that score what the agent submits, by executing its detection queries in the telemetry store and judging its outputs against the answer key. Rather than combining them into one more overall reward, we show that different models lead different stages, which suggests routing each stage of the workflow to the model that performs it best rather than choosing a single model by its overall score. We also report how many tokens each model generates, and find that the share of output spent on thinking does not track the reward.
Abstract (translated)
URL
https://arxiv.org/abs/2610.14180
PDF
