中 EN
登录 注册账户

Academic Intelligence · Curated Daily

探索全球前沿学术脉络

AcademicHub 汇聚顶级期刊与预印本平台的实时文献。定制您的专属科研雷达，利用大语言模型自动生成交叉领域文献分析简报。

登录 注册账户
01.
arXiv (quant-ph) 2026-06-24

Mode-selective nonlinear interference for high-brightness and high-purity fiber-coupled SPDC sources

作者:
Carlos Sevilla-Guti\'errez↗Purujit Singh Chauhan↗Varun Raj Kaipalath↗Fabian Steinlechner↗

arXiv:2606.23836v1 Announce Type: new Abstract: Single-mode-fiber-coupled spontaneous parametric down-conversion (SPDC) sources are a key resource for photonic quantum technologies, but in single-crystal geometries brightness, heralding efficiency, and spectral purity remain constrained by intrinsic trade-offs. Here, we show how nonlinear interference in a cascaded two-crystal type-II SPDC source can be used to engineer the modal structure of SPDC emission, improving the brightness–heralding-efficiency trade-off by more than one order of magnitude beyond the single-crystal limit. We further demonstrate two routes to near-unity spectral purity while retaining high brightness and/or heralding efficiency, even with standard periodically poled crystals, and study the additional advantages of aperiodic poling with Gaussian phase matching. Using a spectrally resolved Laguerre–Gauss modal decomposition, we show that these improvements arise from mode-selective interference of spatial-spectral SPDC modes within the nonlinear interferometer. We experimentally validate the model through sum-frequency-generation measurements of the spatial-spectral state.

阅读与讨论 → 访问原文 →
03.
bioRxiv (Bioinfo) 2026-06-24

trAIt: Species-by-Trait Data Retrieval using Large Language Models

作者:
Balaji↗Martinson↗K. A↗Schellenberger↗J. S↗Koley↗Inman↗C. M↗Hofmann↗H. A↗Young↗R. L↗…

Biological research often requires information about species' traits. Manual literature collation can be time-consuming and miss parts of the literature. To address this gap, we developed trAIt, a publicly available software for the retrieval of characteristics of species from scientific literature catalogued in the Europe PubMed Central (PubMed) database. trAIt provides a graphical user interface in which users specify species and characteristics of interest. Leveraging a large language model (LLM), trAIt retrieves relevant papers, combines their content through a consensus-based summarization model, and outputs a species-by-characteristic table. For a case study involving frog species, trAIt recovered 47.1% of trait-species combinations in 2.75 hours, while an expert curator independently recovered 62.4% over months. The consensus-based summarization substantially aids accuracy compared to single-source extraction. Across three case studies of vertebrate taxa, an expert confirmed the accuracy of 70.9% of trait-species entries recovered by trAIt. We observed considerable variation across taxa in trAIt's accuracy, which is possibly due to heterogeneity in open-access literature availability and inconsistencies in species and trait terminology. In sum, our analysis suggests that LLM-based tools can accelerate biological data synthesis but should be used to support domain experts' research, rather than replace their judgment.

阅读与讨论 → 访问原文 →
04.
arXiv (CS.AI) 2026-06-16

Is Code Better Than Language for Algorithmic Reasoning

作者:
Terry Tong↗Yu Feng↗Surbhi Goel↗Dan Roth↗

arXiv:2606.15589v1 Announce Type: cross Abstract: For tool-augmented language models, comparing natural-language reasoning with code-execution pipelines is difficult because the comparison changes both the intermediate representation and the execution mechanism. We separate these factors with an intermediate intervention: the model expresses its reasoning as executable code, and the language model simulates that code in context to produce an answer. On a 40-task verifiable algorithmic benchmark, deterministic code execution outperforms natural-language reasoning by +31.6pp. We observe that the intermediate intervention is not meaningfully different from natural-language reasoning (+0.15pp). These results suggest that, in our evaluated setting, changing the intermediate representation alone does not explain the tool-use advantage, providing evidence for the performance gains requiring reliable external execution. We formalize this intuition with a simple statistical decision-theoretic model that characterizes when execution dominates end-to-end risk in our disentangled trace-generation/execution regime. We validate our theory using a reconstruction intervention that leverages a proxy language model to infer natural-language reasoning traces from code representations, recovering performance comparable to the original natural-language reasoning pipeline. All experiments are at https://github.com/TerryTong-Git/ToolProj.

阅读与讨论 → 访问原文 →
05.
arXiv (CS.CL) 2026-06-19

Target-Side Paraphrase Augmentation for Sign Language Translation with Large Language Models

作者:
Pedro Dal Bianco↗Jean Paul Nunes Reinhold↗Oscar Stanchi↗Facundo Quiroga↗Franco Ronchetti↗Ulisses Brisolara Corr\^ea↗

Sign language translation (SLT) remains constrained by the limited availability of paired sign-video/text corpora and by the heavy-tailed vocabularies typical of real-world datasets. We study a target-side augmentation strategy in which a large language model (LLM) generates controlled paraphrase variants of the reference spoken-language sentence while the sign input remains unchanged. Concretely, we use GPT-4o to produce semantically faithful variants of the training targets and train a Signformer-style pose-based Transformer under a two-stage schedule: pre-training on the augmented corpus followed by fine-tuning on the original references. We evaluate this strategy on three datasets that span complementary challenges: PHOENIX14T (German Sign Language), a real-world corpus with moderate lexical diversity; the Greek Sign Language Dataset with highly controlled, repetitive recordings; and LSA-T (Argentinian Sign Language), a naturalistic corpus with a large vocabulary and severe long-tail sparsity. This range allows us to characterize precisely when and why target-side augmentation is beneficial. On PHOENIX14T, augmentation improves BLEU-4 from 9.56 to 10.33, demonstrating that paraphrastic exposure helps the decoder generalize beyond memorized reference phrasing. The near-saturated GSL baseline and the extremely sparse LSA-T setting reveal the limits of the approach: in both cases, single-reference lexical overlap metrics are insufficient to capture the full picture, motivating a complementary semantic evaluation. To our knowledge, this is the first study to examine LLM-generated target-side paraphrases as an augmentation mechanism for SLT, and the first to apply an LLM-as-a-Judge evaluation protocol to SLT. This complementary evaluation reveals gains in semantic fidelity that lexical overlap metrics understate.

阅读与讨论 → 访问原文 →
06.
arXiv (CS.CV) 2026-06-16

Trusting Right Predictions for Wrong Reasons: A LIME Based Analysis of Deep Learning Interpretability in Lung Cancer Diagnosis

作者:
Samarpan Poudel↗Vladislav D Veksler↗

Lung cancer is the leading cause of cancer-related mortality, with approximately 2.5 million new cases and 1.8 million deaths annually, making reliable diagnosis a clinical priority. Although deep learning models have achieved strong performance in lung cancer classification, evaluation has largely focused on predictive accuracy, leaving their decision-making processes insufficiently examined. This study compares three architecturally distinct models: a Convolutional Neural Network (CNN), a pretrained ResNet50, and a Vision Transformer (ViT), trained on the IQ-OTH/NCCD lung cancer CT dataset. Local Interpretable Model-Agnostic Explanations (LIME) were applied to investigate model reasoning. In addition to standard performance metrics, a dual-correlation framework was introduced to measure both prediction agreement and explanation agreement across model pairs. All three models achieved strong classification performance, with ResNet50 attaining 98.61% accuracy, CNN 97.91%, and ViT 93.75%, while all achieved ROC-AUC scores of 0.99. Prediction correlations exceeded 0.99 across all model pairs, indicating highly consistent outputs. However, LIME explanation correlations remained below 0.26, revealing substantial differences in the image regions used to reach those predictions. Analysis of misclassified samples further identified a consistent spatial pattern: incorrect predictions were associated with attention outside the lung parenchyma, whereas correct predictions focused primarily within lung regions. These findings demonstrate that prediction agreement is a poor proxy for reasoning consistency, and that interpretability evaluation must be treated as an independent validation criterion alongside predictive performance in clinical AI systems.

阅读与讨论 → 访问原文 →
07.
arXiv (CS.CL) 2026-06-16

AmchiBias: Measuring Stereotypical Bias in Goan Identity Groups with a Minimal Pair Dataset in English and Konkani

作者:
Michelle Barbosa↗Sebastian Pad\'o↗Franziska Weeber↗

Socio-cultural stereotypical bias is an important consideration in the development and deployment of NLP systems. It is however often considered only at the national level, despite rich subnational socio-cultural structures. We present AmchiBias, the first benchmark for measuring socio-cultural stereotypical bias for the Indian state of Goa with its unique historically multicultural setting. It covers various Goan identity groups and comprises 313 minimal pairs across eight sociodemographic dimensions in both English and Devanagari Konkani. We then evaluate stereotypical bias in five multilingual encoder models on this benchmark. We find near-chance scores in Konkani, reflecting language incompetence for general multilingual models and a lack of Goan cultural competence for Indian language models. Queried in English, models with a stronger Indian language coverage show higher bias for pan-Indian groups than hyperlocal Goan groups. This suggests the English signal reflects pan-Indian pretraining associations rather than genuine Goan cultural knowledge. Our findings highlight a critical gap in low-resource multilingual NLP evaluation for hyperlocal community identities.

阅读与讨论 → 访问原文 →
08.
medRxiv (Medicine) 2026-06-17

Identifying anaphylaxis using weakly-supervised prediction models and natural language processing

作者:
Williamson↗B. D↗Cronkite↗D. J↗Yu↗Ramaprasan↗Fuller↗Covey↗Kiniry↗Park↗Winter↗Whitaker↗…

Objectives Scalable computable phenotyping algorithms are critical for conducting high-throughput disease-outcome research in large, distributed-data electronic health record (EHR) and claims data settings. We developed and evaluated a claims- and EHR-based computable phenotyping algorithm for anaphylaxis, a rare acute condition that is challenging to accurately identify using claims data alone. Materials and Methods Potential anaphylaxis events came from two healthcare systems (Kaiser Permanente Washington [KPWA] and Vanderbilt University Medical Center [VUMC]). We engineered features from clinical text using automated natural language processing (NLP) methods. We then developed a phenotyping algorithm using four NLP- and diagnosis code-based silver labels (proxies for the gold-standard labels). Gold-standard abstracted outcomes were used to evaluate algorithm performance. Results The largest area under the receiver operating characteristic curve (AUC) was 0.931 for an NLP-based silver-label model at KPWA. Depending on the model and healthcare system site, positive predictive value (PPV) and sensitivity at the threshold of predicted probability that maximized F1 score ranged from 0.52 to 0.77 (PPV) and 0.78 to 1 (sensitivity). Discussion NLP-based silver-label models had large AUC at KPWA but not at VUMC. This may be because clinical text at KPWA is only available for outpatient encounters and secure messaging. High sensitivity for identifying anaphylaxis can be obtained using our best-performing models. Conclusion The best-performing models had better PPV and sensitivity tradeoffs than prior bespoke anaphylaxis models with costly, manually curated features. The simplicity of the approach compared to traditional phenotyping methods allows it to be deployed easily at multiple health care systems.

阅读与讨论 → 访问原文 →
09.
arXiv (CS.CL) 2026-06-24

Measuring User's Mental Models of Speech Translation in Human-AI Collaboration

作者:
HyoJung Han↗Nishant Balepur↗Jordan Boyd-Graber↗Marine Carpuat↗

Millions of people use machine translation (MT) tools daily, yet little is known about their perception of what systems can and cannot do. This paper studies users' mental models of speech translation systems through a new framework based on cross-lingual question answering, where users either accept MT output or request professional re-translation to answer questions based on the information presented in a foreign language. By analyzing user behavior and accuracy trends across varying translation qualities, we examine to what extent they can predict where the system is likely to be wrong, and how this mental model evolves. Users develop stronger mental models with practice, especially when they have some knowledge of the source language, primarily by relying on surface-level error cues. Moreover, providing speech transcriptions can help users develop better mental models. Our results show the promise of cross-lingual question answering as a downstream task for studying MT mental models and advancing our understanding of human-AI collaboration.

阅读与讨论 → 访问原文 →
10.
Nature (Science) 2026-06-24

Fourier pixels for bidirectional light control

作者:
Yannik M. Glauser↗

Digital cameras1 and displays2 use picture elements (pixels3) that perform a single function: detecting or emitting light intensity. To exploit the full information content of electromagnetic waves, more advanced elements are required. This has driven the development of multifunctional components that, for example, simultaneously detect and emit intensity4,5 or extract intensity and spectral information6–8. However, no pixel exists that both senses and generates optical wavefronts with full control over amplitude, phase and polarization, limiting bidirectional control and feedback of sophisticated light fields. Here we present a route to such pixels by demonstrating a versatile platform of miniaturized diffractive elements based on Fourier optics9. We use plasmonic surface waves10, which propagate coherently11 and efficiently12–15 across metallic surfaces. When these plasmons are launched towards wavy microstructures16 designed with simple Fourier analysis, arbitrary and background-free optical wavefronts are generated. Conversely, incoming light can be sensed, and its amplitude, phase and polarization can be fully characterized. By combining or superposing several such components, we create multifunctional ‘Fourier pixels’ that provide compact and accurate control over the optical field. Our approach, which we extend to photonic waveguide modes, establishes a scalable, universal architecture for vectorially programmable pixels with applications in adaptive optics17,18, holographic displays19–21, optical communication22,23 and quantum information processing24. A versatile platform of miniaturized Fourier-optics-based diffractive elements enables multifunctional pixels that fully control and sense the amplitude, phase and polarization of optical wavefronts for advanced photonic applications.

阅读与讨论 → 访问原文 →
11.
medRxiv (Medicine) 2026-06-16

Development of a symptom-based severity score anchored to health-related quality of life post-COVID-19 within the population-based EPILOC cohorts

作者:
Peter↗R. S↗Sedelmaier↗Nieters↗Schilling↗Matits↗Goepel↗Merle↗Steinacker↗J. M↗Kern↗W. V↗…

Purpose Because simple symptom counts treat all symptoms as equally important and may not adequately capture the HRQoL impact of heterogeneous post-COVID-19 symptoms, we aimed to develop an HRQoL-anchored symptom severity score providing an interpretable measure of post-COVID-19 disease burden. Methods Baseline data from the population-based EPILOC and EPILOC Omicron surveys (adults aged 18-65 years) were used to develop a symptom-based severity score anchored to physical and mental HRQoL assessed with the SF-12. A two-stage modelling approach was applied to identify HRQoL-relevant symptoms and to derive symptom-specific weights for physical and mental component scores, incorporating 30 ordinal symptom severity variables. Symptom-specific weights were extracted to compute physical, mental, and composite severity scores. Score interpretation was examined using external reference measures, including EPILOC case status, self-reported health recovery, and functional consequences. Results A total of 19,004 participants (mean age 44.3 years, 59.6% female) were included. Sixteen symptoms contributed to the physical and eleven to the mental HRQoL score, with a limited subset accounting for most of the HRQoL loss. Severity scores were heavily right-skewed, with 50.6% of participants showing no measurable HRQoL impairment. Higher scores correlated with lower self-reported recovery, and increased probability of rehabilitation use and health-related changes in working time, supporting convergent and criterion-related validity. Conclusions This study introduces a transparent, HRQoL-anchored symptom severity score that measures graded post-COVID-19 burden beyond simple symptom counts. The score may be particularly suited for longitudinal assessment of recovery trajectories.

阅读与讨论 → 访问原文 →
12.
Nature (Science) 2026-06-16

Mathematicians are developing rules for AI use — other fields should follow

作者: 未知作者

The mathematics community is right to call for transparency, integrity and fairness to be protected when AI tools are used. Researchers in other disciplines could learn from this approach. The mathematics community is right to call for transparency, integrity and fairness to be protected when AI tools are used. Researchers in other disciplines could learn from this approach.

阅读与讨论 → 访问原文 →
13.
arXiv (CS.AI) 2026-06-18

Scaling Learning-based AEB with Massive Unlabeled Data

作者:
Xiangyu Wang↗Yang Zhan↗Mengxiang Hao↗Chuanchuan Zhong↗Yansong Jia↗Junjie Zhang↗Yu Han↗Xin Jiang↗Zhen Cao↗Ying Wang↗Yulun Song↗Zhitao Xu↗…

arXiv:2606.18864v1 Announce Type: cross Abstract: This paper studies how to scale learning-based automatic emergency braking (AEB) with massive unlabeled fleet data under production constraints. Our approach is based on meta-feedback semi-supervised learning (MF-SSL), where a teacher generates pseudo labels for unlabeled driving data and is updated using a small labeled anchor set as safety-critical feedback. In production, anchor ambiguity and labeled-unlabeled mismatch can amplify systematic pseudo-label errors, leading to spurious triggers. We propose a stabilized MF-SSL framework with (i) Noise-Aware Decoupling, which removes ambiguity-prone anchors from the teacher's supervised update path, and (ii) kinematics-gated pseudo-labeling with a teacher conflict penalty to suppress mismatch-induced risk hallucinations on unlabeled data while maintaining broad coverage. Extensive experiments show consistent gains as unlabeled data scale from 1M to 1B windows, improving safety while keeping comfort stable. The 1B-trained student model is deployed to hundreds of thousands of vehicles and validated over \$10^9$ km of driving, achieving a positive-to-false activation ratio exceeding 100:1 and a 35% improvement in accident-free driving mileage over a production rule-only baseline.

阅读与讨论 → 访问原文 →
14.
arXiv (CS.AI) 2026-06-11

Time-Series Foundation Model Embeddings for Remaining Useful Life Estimation

作者:
Amir El-Ghoussani↗Michele De Vita↗Ronald Naumann↗Valiseios Belagiannis↗

arXiv:2606.11990v1 Announce Type: cross Abstract: Remaining Useful Life (RUL) prediction is essential for industrial predictive maintenance, yet many learning-based approaches rely on extensive feature engineering or large labeled datasets to train task-specific sequence models. In this work, we introduce a lightweight learning approach, in which we leverage a frozen pretrained time-series foundation model (TSFM) and combine it with a small regression head for RUL estimation from multivariate sensor streams. More specifically, we use Chronos-2 as a frozen backbone to extract context window features and train a lightweight regression neural network for RUL prediction. Experiments on real-world industrial sensor data from two device types show that Chronos-2 features consistently improve over recurrent, convolutional, Transformer-based, and gradient-boosting baselines under the same preprocessing and evaluation protocol. We further analyze the impact of context length and find that performance improves significantly with longer histories, indicating that TSFM representation offer a practical and data-efficient alternative for RUL estimation in industrial settings.

阅读与讨论 → 访问原文 →
16.
arXiv (quant-ph) 2026-06-24

Rotational Vacuum Friction of Nonabsorbing Particles

作者:
F. Javier Garc\'ia de Abajo↗Alejandro Manjavacas↗

arXiv:2606.24723v1 Announce Type: new Abstract: A nonabsorbing particle rotating in vacuum can lose angular momentum only by converting mechanical energy into electromagnetic radiation. Here, we develop a quantum theory of rotational vacuum friction for small lossless particles and show that axial symmetry qualitatively changes the leading dissipation channel. At zero temperature, the frictional torque scales as $M\propto\Omega^7$ with rotation frequency $\ Omega$ in anisotropic particles due to the emission of correlated photon pairs whose frequencies sum to $2\Omega$, while a contribution to the torque linear in $\ Omega$ is found at finite temperature. In contrast, axisymmetric particles are protected against photon-assisted friction regardless of temperature.

阅读与讨论 → 访问原文 →
17.
arXiv (CS.LG) 2026-06-12

Scalable anomaly detection via a univariate Christoffel function

作者:
Florian Grivet↗Didier Henrion↗Jean-Bernard Lasserre↗Louise Trav\'e-Massuy\`es↗

arXiv:2606.12483v1 Announce Type: new Abstract: Anomaly detection plays a critical role in identifying unusual patterns across domains such as fraud detection, network intrusion, and system fault diagnosis. Recently, Christoffel function-based methods, rooted in polynomial optimization, have emerged as promising alternatives to deep learning due to their strong mathematical foundations and computational frugality. However, their practical applicability is hindered by the need to invert a matrix whose size grows exponentially with the data dimension, rendering the method intractable even for moderate-dimensional datasets. This paper addresses the dimensionality limitations of Christoffel function-based anomaly detection while preserving its key theoretical properties, i.e., the on-off support dichotomy behavior and the accurate support shape capture. We introduce UCF, a univariate Christoffel function which is based on the squared distance between the query point and the support points. Extensive experiments on the ADBench benchmark demonstrate that UCF consistently outperforms 14 state-of-the-art baselines in terms of Average Precision. By resolving the scalability bottleneck of the Christoffel Function, this work expands the toolkit of anomaly detection methods with a robust, theoretically grounded, and universally applicable approach.

阅读与讨论 → 访问原文 →
18.
arXiv (CS.AI) 2026-06-11

FreeBridge: Variational Schrödinger Bridges for Cellular Transition Dynamics

作者:
Xurui Wang↗Qin Ren↗Jun Ma↗Haibin Ling↗Chenyu You↗

arXiv:2606.11286v1 Announce Type: cross Abstract: High-content imaging assays quantify cellular responses to chemical and genetic perturbations, yet continuous trajectories of individual cells are unobservable because cells are chemically fixed at acquisition. Perturbation modeling therefore reduces to inferring stochastic transport between control and treated populations observed only as separate marginals. While recent generative models achieve strong end-point alignment, boundary consistency does not determine intermediate evolution: multiple stochastic processes may connect identical marginals while traversing regions unsupported by observed single-cell morphologies. We introduce FreeBridge, a Schrödinger Bridge formulation for single-cell transition modeling under endpoint-only supervision. FreeBridge defines atomic states as instance-segmented single-cell representations, establishing a fixed cellular manifold, and learns stochastic transport constrained within this geometry via empirical latent support regularization. Across BBBC021, RxRx1, and JUMP, FreeBridge maintains competitive or improved endpoint fidelity and mechanism-of-action retention under a unified evaluation protocol; on BBBC021, it further reduces intermediate support violations. These findings highlight the importance of geometric grounding for biologically interpretable perturbation dynamics. Project page: https://y-research-sbu.github.io/FreeBridge/.

阅读与讨论 → 访问原文 →
19.
arXiv (CS.CL) 2026-06-17

Improving low-resource ASR using bilingual fine-tuning with language identification: a cross-linguistic evaluation

作者:
Reihaneh Amooie↗Yun Hao↗Wietse de Vries↗Jelske Dijkstra↗Matt Coler↗Martijn Wieling↗

This study explores how bilingual fine-tuning affects automatic speech recognition (ASR) in low-resource languages. We evaluate this method across nine linguistically and geographically diverse language pairs, covering a range of language families and writing systems. To distinguish the two languages, during training, we pre-pend each input text with a language identification token. At inference, the model jointly predicts both the language and transcription from the speech input alone. As texts for which the language is incorrectly determined show low ASR performance, we also conduct a follow-up experiment in which the language identification token is provided both during training and inference. Our results show that bilingual fine-tuning can be beneficial when language identification accuracy is high, and that in cases where language identification performance is low, including the language identification token at inference helps to improve ASR performance.

阅读与讨论 → 访问原文 →
22.
Nature (Science) 2026-06-17

The ancestors of eukaryotic cells contained a mix of genes from various microbes

作者: 未知作者

Reconstruction of the ancestral gene repertoire of eukaryotic cells reveals traces of a series of close, long-term interactions with diverse microorganisms, and a role of viruses in gene exchange. The findings challenge the view that eukaryotic cells evolved from a simple merger of just two organisms. A series of gene-transfer events might have taken place in complex microbial communities.

阅读与讨论 → 访问原文 →
23.
arXiv (CS.CV) 2026-06-19

Spectral Query-Key Product Weight Steering for Training-Free VLM Hallucination Mitigation

作者:
Karn Tiwari↗Varnith Chordia↗Prathosh A P↗

Vision-language models (VLMs) often generate fluent but visually unsupported descriptions, especially by mentioning objects absent from the image. We propose QK Product Steering, a data-free, training-free, and zero-inference-cost weight edit for reducing object hallucination. The method directly edits the per-head query-key product, the operator that produces pre-softmax attention logits, by suppressing a small number of dominant singular modes in selected middle layers. The edited product is then mapped back to the query weights through a closed-form query-only update while keeping shared key weights fixed, making the edit compatible with grouped-query attention. We further decompose the QK product into symmetric and antisymmetric components to distinguish mutual content-similarity patterns from directional attention patterns. Across three GQA-based VLMs, QK Product Steering achieves an average relative CHAIR$_s$ reduction of $4.0\%$, while matched random-mode controls show negligible change. Interpretability ablations show that the hallucination signal is specific to dominant QK modes and is primarily localized to the symmetric mutual-attention channel. Overall, QK Product Steering offers a simple alternative to decoding-time mitigation, requiring no additional data, fine-tuning, or inference-time overhead while largely preserving general multimodal capability.

阅读与讨论 → 访问原文 →
24.
arXiv (CS.CV) 2026-06-24

REALM: A Unified Red-Teaming Benchmark for Physical-World VLMs

作者:
Yifei Zhao↗Qian Lou↗Mengxin Zheng↗

Vision-language models (VLMs) are increasingly used as perception-reasoning backbones for embodied intelligence in safety-critical physical systems, where perception or reasoning errors can lead to unsafe decisions or actions. Although many red-teaming methods have been developed to probe VLM vulnerabilities, their evaluation remains fragmented across datasets, metrics, and threat models, making direct comparison difficult and obscuring whether observed differences arise from stronger attacks, more vulnerable models, or incompatible evaluation settings. Existing chatbot-centric red-teaming benchmarks mainly standardize jailbreak and content-safety evaluation, but they do not systematically capture physically grounded functional failures or cover red-teaming methods that target physical-world VLMs. This raises the key challenge of comparing diverse attack methods under a unified protocol while targeting the same scenario-specific failures. We introduce REALM, to our knowledge the first unified red-teaming benchmark for physical-world VLMs. REALM integrates 12 red-teaming methods, 3 model-agnostic defenses, and 13 VLMs under a practical black-box threat model with shared datasets and metrics. To align adversarial objectives across attack families, REALM introduces an agentic target-generation pipeline that constructs shared, scenario-specific, and physically grounded attack objectives for each scene, enabling fair comparison of diverse red-teaming methods under aligned adversarial goals. Our evaluation shows that text and typographic injection attacks induce the most failures, multimodal co-optimization yields the strongest visual-perturbation transfer, single-pass attacks approach iterative methods at much lower cost, and model scale alone does not confer adversarial robustness. Code is available at https://github.com/UCF-ML-Research/REALM.

阅读与讨论 → 访问原文 →
25.
arXiv (CS.CL) 2026-06-16

LLM Judges Have Dark Current: A Psychometric Datasheet for LLM-as-a-Judge Evaluation

作者:
Hiroyasu Usami↗Keisuke Hara↗Ayato Tsuboi↗Naohiko Matsuda↗

LLM-as-a-judge systems are now routinely used for open-ended model evaluation, where human preference annotation is costly, slow, and difficult to reproduce. Yet these judges are often reported as scalar accuracy, win-rate, or agreement devices. We argue that a judge should instead be reported as a measurement instrument. We introduce a Judge Datasheet protocol that measures dark current under true-vacuum inputs, stable cross-sensitivity to same-quality surface variation, positional false preference, target sensitivity on a controlled quality ladder, and the criterion or operating point induced by tie instructions. The direction-stability decomposition reveals that apparent Delta0 preference can be stable surface response or disguised position bias. In a three-judge open-weight case study, Llama-3.1-8B shows high dark current and presentation-conflicted Delta0 behavior, Qwen2.5-14B is vacuum-clean and target-sensitive but mixes stable and positional over-discrimination, and Qwen2.5-32B is vacuum-clean with low stable cross-sensitivity and low positional false preference. A strict tie criterion eliminates Qwen32B Delta0 false preference but absorbs marginal Delta1 target signals into ties while preserving Delta5 sensitivity. The results show that prompting moves the criterion, not the resolution. We do not claim that the downstream mechanism hypothesis that motivated this work is confirmed; the contribution is a metrological protocol for measuring the measuring device before downstream claims are made.

阅读与讨论 → 访问原文 →