Publications

2026

• Why Question Answering Modality Matters in Visualization Studies and Tests
  
  • Cabouat Anne-Flore
  • Ge Lily W
  • Isenberg Tobias
  • Huron Samuel
  • Kay Matthew
  • Isenberg Petra
  , 2026. In this position paper, we argue that answer modality choices can affect the validity of visualization evaluations. Visualization evaluation studies frequently rely on specific answer modalities such as multiple-choice questions, open-ended numerical estimates, verbal explanations, or spatial interactions. Yet, answer modality is often treated as a neutral reporting mechanism rather than as a factor that may influence reasoning strategies, representational transformations, and ultimately measured performance and the conclusions researchers can draw from their observations. We discuss two broad classes of modality effects on participant performance: changes in participants' task-solving strategies and changes induced by representational transformations between the visual stimulus and the expected response format. We further argue that answer modality determines what evidence about participants' reasoning is preserved for analysis. We ground our reflections in observations from prior empirical work and propose practical guidance for reasoning about answer modality as part of visualization evaluation design.
• Data Processing Inequalities for the Quantum Guesswork
  
  • Mejdoub Ilyass
  • Béguinot Julien
  • Rioul Olivier
  , 2026. <div><p>Quantum guesswork has emerged as a crucial measure in assessing the distinguishability of nonorthogonal quantum states. In this paper, we review and compare previous definitions of the quantum guesswork, and establish the pre and post data processing inequalities (DPIs) that are relevant for the guessing problem. We also derive improved lower and upper bounds on the quantum guesswork in terms of the Shannon entropy and Holevo's information.</p></div>
• Secrecy Metrics over Random Gaussian SIMOME and MIMOME Channels
  
  • Leroy Clément
  • Arbi Tarak
  • Geller Benoit
  • Rioul Olivier
  , 2026. <div><p>The achievable secrecy of time-invariant wireless channels is typically characterized by the secrecy capacity of the communication system, whose mathematical expression is well established. However, in most practical scenarios, channels cannot be assumed to remain constant or to be perfectly known at both the transmitter and the legitimate receiver. In such cases, no closed-form expression for the secrecy capacity is available.</p><p>To address this, several metrics are studied, that characterize secrecy in random channel environments: Ergodic Secrecy Rate (ESR), Ergodic Positive Secrecy Rate (EPSR), and Secrecy Outage Probability (SOP). We derive the optimal values of these metrics for SIMOME channels (Single-Input, Multiple-Output, Multiple-Eavesdropper) and various MIMOME (Multiple-Input, Multiple-Output, Multiple-Eavesdropper) scenarios.</p></div>
• Pseudonym Certificate Revocation in V2X: Liability, Scalability and Latency. A Quantitative Analysis and a Roadmap toward a Standardizable Solution
  
  • Grund Jean-Jacques
  • Khatoun Rida
  • Pujolle Guy
  • Serhrouchni Ahmed
  , 2026. <div><p>The security of V2X communications relies on pseudonym certificates that sign a very large volume of cooperative messages (CAM, DENM, BSM, SRM) at rates of up to 10 Hz, in environments where connectivity is intermittent. Under these conditions, the revocation of these certificates is the most challenging link of the trust chain: it couples three frequently antagonistic requirements -liability (the ability to attribute and rapidly exclude a compromised device without breaking conditional privacy), scalability (the effects of pseudonym inflation and of the cumulative volume of revocations), and latency (a budget of a few tens to a few hundreds of milliseconds per application decision). This paper provides a quantitative analysis of the state of the art, structured along the three stages of the revocation process (resolution, distribution, use). We formalize six requirements R1-R6 and a homogeneous sizing model, and we show, using explicit European-scale deployment figures, that neither classical X.509 PKI nor any individual family of existing schemes (CRL and its optimizations, probabilistic filters, SCMS linkage, short-lived certificates, Merkle structures, online and symmetric approaches, group signatures, distributed ledgers) simultaneously satisfies the six requirements. We identify eight structural gaps and translate them into a roadmap of eight objectives O1-O8. This paper constitutes Part I of a two-part contribution: it establishes the analytical foundation -requirements, quantitative sizing, structural gaps, and a specification-level roadmap -while the design of a compliant mechanism, its security analysis, and its experimental evaluation are presented in a dedicated companion paper (Part II).</p></div>
• Physical Layer Security with Frequency-Hopping and Finite-Alphabet Inputs
  
  • Leroy Clément
  • Arbi Tarak
  • Geller Benoit
  • Rioul Olivier
  , 2026. <div><p>This paper investigates the physical layer security of frequency-hopping (FH) transmission schemes when the hopping sequence is publicly known. We consider a single-antenna transmitter communicating with a legitimate receiver over a channel estimated with a known error distribution, in the presence of a passive eavesdropper whose channel statistics are available at the transmitter. We show that, when the number of hopping frequencies is sufficiently large, the instantaneous channel fluctuations induced by FH transmissions can be effectively averaged out, and the secrecy performance is accurately predicted by the ergodic secrecy rate (ESR). The ESR is maximized by selecting the optimal modulation order from a set of admissible finite constellations. We derive a closed-form condition on the optimal modulation order, which only depends on the tail probabilities of the legitimate and eavesdropper signal-to-noise ratio (SNR) distributions, and we propose an efficient algorithm for its computation. In particular, we demonstrate that a positive secrecy rate can be achieved even when the eavesdropper has a higher average SNR than the legitimate receiver, provided that the spread of the eavesdropper SNR distribution is sufficiently large. Numerical results validate the proposed framework under a variety of channel models.</p></div>
• Not In My Git Yard: Catching Backdoors at Commit and Release Time
  
  • Kokkonis Dimitri
  • Marcozzi Michaël
  • Zacchiroli Stefano
  , 2026. Code-level backdoors-stealthy code changes that grant hidden privileges via secret triggers-pose a persistent threat to opensource software. Known attempts to inject such backdoors into widely used projects through malicious commits, tampered release packages, or compromised third-party dependencies, were stopped only by luck and manual review. Existing Continuous Integration (CI) pipelines cannot detect these attacks, and downstream binary analysis tools require substantial manual effort. In this work, we present Lily, an automated approach that strengthens open-source development and release processes against backdoor injection. Lily integrates a backdoor detection mechanism into (1) CI pipelines to block malicious commits, and (2) release vetting workflows to prevent tampered releases or compromised dependencies from entering large ecosystems, such as Linux distributions. Lily offers two key contributions. First, it enhances CI-compatible fuzzing with the capability to detect triggers of suspicious behavior based on historical and current software executions. This enables fast, precise backdoor detection suitable for both CI and update validation workflows. Second, it combines code change analysis with fuzzing data to precisely point maintainers to backdoor-revealing code regions, even when release updates modify millions of lines of code. We also outline five strategies attackers could use to evade Lily, and evaluate corresponding defenses. Our experiments across hundreds of benign and backdoored commits and releases show that Lily achieves high detection accuracy with low false alarm rates, reliably identifies malicious code, resists adversarial attempts, and would have prevented real-world backdoor incidents. (10.1145/3832783.3834352)
  DOI : 10.1145/3832783.3834352
• RISCGuard: RISC-V Static Analysis Framework for Detecting Microarchitectural Vulnerabilities
  
  • Khan Mahreen
  • Mushtaq Maria
  • Pacalet Renaud
  • Apvrille Ludovic
  , 2026. <div><p>The proliferation of RISC-V processors in security-critical applications has raised concerns about side-channel vulnerabilities, particularly those arising from speculative execution and cache timing attacks. While software-based mitigation strategies exist, automated detection of vulnerable code patterns remains challenging. This paper presents RISCGuard, a novel static analysis framework specifically designed to identify side-channel vulnerabilities in RISC-V assembly code. RISC-Guard employs sophisticated pattern matching, data-flow analysis, and taint tracking to detect ten distinct vulnerability classes, including Spectre variants (V1, V2, V4), cache timing attacks (Flush+Reload, Prime+Probe), cache conflict patterns, timing side-channels, password timing vulnerabilities, secret-dependent control flow, and unaligned memory access patterns. We evaluate RISCGuard on a comprehensive dataset comprising 100 benign sequences, 125 known vulnerable patterns, and 100,000 randomly generated instruction sequences. Our experimental results demonstrate an overall accuracy of 99.6% across all test cases, with low false-positive and false-negative rates. The framework generates detailed vulnerability reports that include risk assessments ranging from critical to low, confidence scores averaging 0.89, and mitigation recommendations. Our paper addresses a crucial gap in RISC-V security tooling and provides developers with an automated solution for identifying and mitigating side-channel vulnerabilities before deployment.</p></div>
• SPS: Simulate. Predict. Secure; Pre-Silicon Validation of Embedded Kernels via ML-Enhanced Cycle-Accurate Simulation and Anomaly Detection
  
  • Awais Muhammad
  • Mushtaq Maria
  • Naviner Lirida
  • Haj Jawad
  • Bruguier Florent
  , 2026, 14th. Pre-silicon validation of embedded processors demands simulation frameworks that simultaneously capture microarchitectural timing, peripheral behavior, and firmware interactions, requirements that functional emulators and analytical models fail to meet. This paper introduces SPS (Simulate. Predict. Secure.), a gem5-based framework extended with embedded kernel models, memory hierarchies, and peripheral co-simulation, coupled with machine learning to accelerate design-space exploration and enable microarchitectural anomaly detection. Representative kernels spanning signal processing, control, communication, and lightweight security workloads are executed in Syscall Emulation mode to collect fine-grained statistics including cache activity, bus contention, interrupt latency, and execution cycles. These features train surrogate ML models that reduce exploration time by up to 10× while recovering over 93% of the Pareto front, with energy and performance prediction errors below 3.2% and 2.8% respectively. The framework further incorporates an anomaly detection layer operating on gem5 performance counters, achieving AUC-ROC ≥ 0.89 across six fault classes including interrupt storms, cache thrashing, memory leaks, and security breaches. Experimental results confirm that SPS delivers efficient exploration, accurate prediction, and robust fault detection, establishing a scalable pre-silicon methodology for energy-aware and secure embedded processor design.
• Sword and Shield: Uses and Strategies of LLMs in Navigating Disinformation
  
  • Lim Gionnieve
  • Tan Bryan Cheng Zhengyu
  • Sim Kellie Yu Hui
  • Shi Weiyan
  • Chew Ming Hui
  • Hee Ming Shan
  • Lee Roy Ka-Wei
  • Perrault Simon
  • Choo Kenny Tsu Wei
  , 2026. The emergence of Large Language Models (LLMs) presents a dual challenge in the fight against disinformation. These powerful tools, capable of generating human-like text at scale, can be weaponised to produce sophisticated disinformation, yet they also hold promise for enhancing mitigation strategies. This paper investigates the complex dynamics between LLMs and disinformation in small, localised settings through a communication game based on online forums, inspired by Werewolf, with 25 participants. We analyse how Disinformers, Moderators, and Users leverage LLMs to advance their goals, revealing both the potential for misuse and combating disinformation. Our findings highlight the varying uses of LLMs depending on the participants' roles and strategies, underscoring the importance of understanding their effectiveness in this context. We conclude by discussing implications for future LLM development and online platform design, advocating for a balanced approach that empowers users and fosters trust while mitigating the risks of LLM-assisted disinformation. (10.1145/3817008)
  DOI : 10.1145/3817008
• Did You Forkget It? Detecting One-Day Vulnerabilities in Open-source Forks With Global History Analysis
  
  • Lefeuvre Romain
  • Reux Charly
  • Zacchiroli Stefano
  • Barais Olivier
  • Combemale Benoit
  , 2026. Tracking vulnerabilities inherited from third-party open-source software is a well-known challenge, often addressed by tracing the threads of dependency information. However, vulnerabilities can also propagate through forking: a code repository forked after the introduction of a vulnerability, but before it is patched, may remain vulnerable long after the vulnerability has been fixed in the initial repository. History analysis approaches are used to track vulnerable software versions at scale. However, such approaches fail to track vulnerabilities in forks, leaving fork maintainers to identify them manually. This paper presents a global history analysis approach to help software developers identify one-day (known but unpatched) vulnerabilities in forked repositories. Leveraging the global graph of public code, as captured by the Software Heritage archive, our approach propagates vulnerability information at the commit level and performs automated impact analysis. Starting from 7162 repositories with vulnerable commits listed in OSV, we propagate vulnerability information to 2.2 million forks. We evaluate our approach by filtering forks with significant user bases whose latest commit is still potentially vulnerable, manually auditing the code, and contacting maintainers for confirmation and responsible disclosure. This process identified 135 high-severity one-day vulnerabilities, achieving a precision of 0.69, with 9 confirmed by maintainers.
• Do Visual Features Improve Other-Initiated Repair Detection? A Dyadic Multimodal Approach
  
  • Ngo Anh
  • Rollet Nicolas
  • Pelachaud Catherine
  • Clavel Chloé
  , 2026. Other-initiated Self-repair, or in short Other-initiated Repair (OIR), is an essential mechanism in conversational interaction, whereby a recipient signals a problem in speaking, hearing, or understanding, prompting the previous speaker to resolve it. In the case of conversational agents, it is essential to accurately identify these repair initiation strategies to address communication breakdowns efficiently. While conversational analysis studies have shown that OIR initiation is accompanied by both verbal and non-verbal signals such as gaze shifts, facial expressions, body postures, and hand gestures, existing computational approaches rely mainly on text and audio. This paper introduces a novel multimodal model for OIR detection and classification, incorporating a set of visual features drawn from conversation analysis. We evaluate our approach on two corpora with distinct languages and interaction settings. Results demonstrate that visual information consistently improves performance over text and audio baselines, and provide insights into cross-modal feature contributions across two corpora. (10.1145/3776574.3831200)
  DOI : 10.1145/3776574.3831200
• A Methodology to Compare Real and LLM-Simulated Multimodal Interactions: The Case of Gender Bias
  
  • Etienne Elodie
  • Ochs Magalie
  • Loubes Jean-Michel
  • Clavel Chloé
  , 2026. Large Language Models (LLMs) are increasingly used to generate multimodal data for building corpora or simulating synthetic behaviours. A key challenge remains the evaluation of generated data. Comparison with real multimodal human interactions may help address this issue, but unified representations and evaluation frameworks are still lacking. In this article, we introduce MAMBI (Multimodal Annotation and Measurements for Bias in Interactions), an open-source tool designed to facilitate systematic comparisons between real and LLM-simulated multimodal human interactions. For this purpose, a standardised enriched multimodal transcript format is proposed, providing a common representation of verbal, prosodic, and non-verbal behaviours. This representation is obtained through an automatic annotation pipeline that transforms raw audio-visual recordings into structured narrative transcripts by combining speech transcript, segmentation, prosodic and facial feature extraction. The pipeline is applied on multiple multimodal corpora, demonstrating its portability across diverse interaction settings and different numbers of participants. Based on this enriched multimodal transcript, a framework is introduced to assess the statistical consistency between real and generated data, focusing on distributional similarity and correlation structure preservation in the prosodic, interactional dynamics, and facial expressions features, with a particular attention on the bias analysis to quantify group-related disparities. As a use case, we present in this article an analysis of \textit{gender} bias on a large \textit{dyadic} interaction corpus. The results show that enforcing distributional constraints substantially improves alignment with real data, whereas structural consistency depends more unevenly on the prompting strategy and is not systematically maximized by correlation-oriented prompting alone. However, all generations amplify gender-related biases, especially in prosodic cues such as pitch and speech rate, and in facial behaviours such as smiling, blinking, and raised cheeks. Overall, the proposed tool facilitates the generation, evaluation, and critical analysis of multimodal interactions, contributing to bridging the gap between real and LLM-simulated data. (10.1145/3776574.3831157)
  DOI : 10.1145/3776574.3831157
• Identification of novel CLK1 inhibitors by computational fragment-based ligand design, co-crystallization, chemical synthesis and structure activity relationships
  
  • Bourg Stéphane
  • Place Matthieu
  • Copin Chloé
  • Chaikuad Apirat
  • Robert Thomas
  • Holzmann Hanna
  • Müller Susanne
  • Bach Stéphane
  • Ruchaud Sandrine
  • Knapp Stefan
  • Buron Frédéric
  • Routier Sylvain
  • Bonnet Pascal
  European Journal of Medicinal Chemistry, Elsevier, 2026, 315, pp.118977. (10.1016/j.ejmech.2026.118977)
  DOI : 10.1016/j.ejmech.2026.118977
• Feature Toggle Dynamics in Large-Scale Systems: Prevalence, Growth, Lifespan, and Benchmarking
  
  • Tërnava Xhevahire
  , 2026. <div><p>Feature toggles enable gradual rollouts and experimentation in software systems, yet often persist beyond their intended lifecycle, accumulating as technical debt. Prior research has examined feature toggle interactions and complexity, but no longitudinal study has quantified how toggles evolve over time across different organizational contexts. We analyse over 4,000 toggle events in Kubernetes (10 MLoC, 8.5 years) and GitLab (5 MLoC, 5 years). We find that feature toggle removals lags behind additions in both systems (by roughly 35% and 13%, respectively), leading to growing toggle inventories. Their lifespan patterns also differ notably, with Kubernetes toggles lasting a median of 734 days versus 185 in GitLab. Then, some feature toggles (1.33% and 0.73%, respectively) exceed all previously observed removal durations, becoming de facto permanent. Building on these findings, we propose a benchmarking framework with five key metrics and their empirically derived threshold zones, enabling practitioners to assess and compare toggle management practices across projects. All scripts and data are publicly available.</p></div> (10.5281/zenodo.18773811)
  DOI : 10.5281/zenodo.18773811
• Where Did the Variability Go? From Vibe Coding to Product Lines by Regeneration
  
  • Tërnava Xhevahire
  , 2026. In vibe coding, an emerging AI-driven paradigm, an LLM generates an entire program from a natural language prompt, but what happens to the variability that traditional software engineering carefully builds into code? To answer this question, we conducted an exploratory analysis on 10 vibe coded C/C++ projects, which suggests that there is near zero in-artifact variability, i.e., at compile- and runtime. All variability decisions are resolved at a single new binding time, generation time, the moment the LLM produces the source code. Rather than treating this as a defect to fix, we propose Variability by Regeneration (VbR), to our knowledge the first product line approach in which the LLM acts as the derivation engine, generating a dedicated, free of dead code binary for each variant from a declarative specification, while a variant dispatcher transparently routes user requests to the matching binary. We formalise VbR, contrast it with classical SPL derivation, and demonstrate its full pipeline on a wc product family. For SPL engineering, variability in AI-generated software belongs in the specification, not in the code. (10.5281/zenodo.20730698)
  DOI : 10.5281/zenodo.20730698
• LMMSE-Based Phase Estimator for Fiber-Optic Coherent Spread-Spectrum FBG-Based Sensing Systems
  
  • Bai Yiyang
  • Awwad Élie
  • Wigger Michèle
  IEEE Photonics Journal, Institute of Electrical and Electronics Engineers (IEEE), 2026. Fiber Bragg grating (FBG)-based sensing enables localized strain and temperature monitoring by tracking the phase changes of the light reflected from FBGs inserted along a fiber. A major source of disturbance in such systems is laser phase noise, and its influence is typically mitigated by means of additional hardware. In this article, we propose to mitigate the influence of laser phase noise by means of digital signal processing. We propose a practical estimator for coherent single-polarization spread-spectrum FBG-based sensing systems, which is motivated by linearizing the complex received signal and approximating the loss function to formulate a minimum mean square error (MMSE) estimation problem. Moreover, a Cramér-Rao lower bound for this FBG-based estimation problem is presented. Our simulation results demonstrate that the performance of our proposed estimator lies close to the Cramér-Rao lower bound, thus indicating that our estimator effectively compensates for phase noise in FBG-based sensing systems.
• Temporal 2G–5G RF-EMF exposure assessment in ten European countries during one year
  
  • van Bladel Han
  • Veludo Adriana Fernandes
  • Loizeau Nicolas
  • Röösli Martin
  • Maule Milena
  • Vecsei Zsuzsanna
  • Molnár Orsolya
  • Vrijkotte Tanja
  • Polanska Kinga
  • Politański Piotr
  • Mamrot Paweł
  • Wang Shanshan
  • Wiart Joe
  • Grellier James
  • Kovalenko Anastasiia
  • Hulls Paige
  • de Vocht Frank
  • Vaupotiĉ Nina
  • Guxens Mònica
  • Joseph Wout
  Science of the Total Environment, Elsevier, 2026, 1047, pp.182037. Due to the rapid evolution and densification of mobile communication networks, there is an increased interest in the long-term assessment of environmental exposure to radiofrequency electromagnetic fields (RF-EMF). While numerous studies have investigated RF-EMF exposure using short-term or spatial measurement campaigns, the number of comprehensive analyses capturing temporal variability across multiple frequency bands and countries remain limited. This pilot study aims to establish a multi-country sensor network to collect data on long-term RF-EMF exposure, including 5G, and to investigate whether temporal trends occur. Twenty frequency-selective sensors were deployed at fixed indoor and outdoor locations in ten European countries, continuously measuring E-field (electric field) strengths in four mobile communication frequency bands (806 MHz, 942 MHz, 1842 MHz, and 3625 MHz) over a period of 17 months with a temporal resolution of 1 s. The raw measurement data were calibrated and analyzed to describe RF-EMF exposure levels, temporal patterns, and variability across environments and spatial characteristics. A strong and recurring diurnal pattern was observed across the four frequency bands. The highest day-night contrasts occurred in the 806 MHz and 1842 MHz bands, where nighttime values decreased by 35.1% and 48.4%, respectively compared to daytime values. Milder contrasts were observed between weekdays and weekends with the most pronounced decrease of 16.6% for the 1842 MHz band. Temporal variability, quantified using the R-factor (i.e. the ratio of the median field strength to the maximum field strength measured during the period of interest), varied substantially across frequency bands and locations, with lower R-factors observed for higher-frequency bands (with a median R-factor of 0.45 for 3625 GHz), indicating a greater variability. Given the scope and duration of this study, it aims to serve as a pilot study for long-term exposure monitoring at fixed sites in multiple countries. (10.1016/j.scitotenv.2026.182037)
  DOI : 10.1016/j.scitotenv.2026.182037
• Statistical wave field theory: Main results
  
  • Badeau Roland
  , 2026. The statistical wave field theory mathematically establishes the statistical laws of the solutions to the wave equation in a bounded domain. It provides the closed-form expressions of the power distribution and the correlations of the wave field jointly over time, frequency and space, which hold at high frequency and after many reflections, in terms of the geometry and the specific admittance of the boundary surface. This paper summarizes the main results of the theory.
• Training-free Uncertainty Guidance for Complex Visual Tasks with MLLMs
  
  • Kim Sanghwan
  • Xiao Rui
  • Alaniz Stephan
  • Xian Yongqin
  • Akata Zeynep
  , 2026. Multimodal Large Language Models (MLLMs) often struggle with fine-grained perception, such as identifying small objects in high-resolution images or detecting key moments in long videos. Existing methods typically rely on complex, task-specific fine-tuning, which reduces generalizability and increases system complexity. In this work, we propose an effective, training-free framework that uses an MLLM's intrinsic uncertainty as proactive guidance. Our core insight is that a model's uncertainty decreases when provided with relevant visual information. We introduce a unified mechanism that scores candidate visual inputs by response uncertainty, enabling the model to autonomously focus on the most informative data. We apply this simple principle to three challenging visual tasks: Visual Search, Long Video Understanding, and Temporal Grounding, allowing off-the-shelf MLLMs to achieve performance competitive with specialized, fine-tuned systems. Our results demonstrate that leveraging intrinsic uncertainty is a powerful strategy for improving fine-grained multimodal performance.
• Look But Don't Touch with Sparse Autoencoders for Unlearning in Diffusion Models
  
  • Cassano Enrico
  • Renzulli Riccardo
  • Ahmed Rayyan
  • Grangetto Marco
  • Alaniz Stephan
  , 2026. Sparse autoencoders (SAEs) have recently been proposed as interpretable tools for concept-level manipulation, under the assumption that isolated features can serve as controllable intervention points. In this work, we systematically evaluate this assumption in the context of object erasure and steering in diffusion models. We show that while SAEs reliably detect and localize semantic concepts within diffusion model activations, direct intervention in their latent space frequently induces out-of-distribution activations, resulting in severe visual artifacts. To disentangle detection from intervention, we use SAE activations purely as semantic detectors to identify image regions containing the target object, and replace those patch embeddings with the ones that do not contain it. This detection-based replacement preserves the diffusion model's activation statistics and produces significantly cleaner erasure results than latent steering. Our findings reveal a fundamental gap between concept detection and concept intervention in diffusion models: monosemantic or sparse features are not inherently suitable as control knobs for steering. These results position SAEs as powerful interpretability tools for analyzing generative models, but highlight important limitations when used for direct manipulation, such as unlearning.
• Single-Query Black-Box Calibration Auditing via Logit Bias
  
  • Plaud Roman
  • Saillenfest Antoine
  • Labeau Matthieu
  • Bonald Thomas
  • Waegeman Willem
  , 2026. Evaluating the calibration of Large Language Models (LLMs) is critical for their safe deployment as zero-shot classifiers. Yet, commercial API providers increasingly hide the continuous output probabilities required by standard calibration metrics. To bypass this opacity, we demonstrate that any LLM API exposing a logit\_bias parameter can be mathematically manipulated to evaluate exact probability thresholds using strictly one query per sample. Leveraging this mechanism, we introduce a novel and provably consistent estimator of the True Calibration Error for binary tasks. Our approach therefore provides an efficient framework for auditing black-box foundation models.
• Towards Trust Calibration in Socially Interactive Agents: Investigating Gendered Multimodal Behaviors Generation with LLMs
  
  • Galland Lucie
  • Clavel Chloé
  • Ochs Magalie
  , 2026. As Socially Interactive Agents (SIAs) become increasingly integrated into daily life, the ability to calibrate user trust to an agent's actual capabilities would help ensure appropriate usage of these agents. In this paper, we explore the capacity of Large Language Models (LLMs) to generate multimodal behaviors (verbal, vocal, gestural, and facial expression modalities) that reflect varying levels of ability and benevolence (A&amp;B), two key dimensions of trustworthiness. We propose a novel method for automatically generating behaviors aligned with specific levels of these traits, a first step towards enabling nuanced and trust-calibrated interactions. By analyzing a large dataset of multimodal transcripts generated by LLMs, we demonstrate that GPT-5.4 is able to produce coherent behavior across different modalities (text, intonation, facial expression, and gesture). Using Random Forest feature importance analysis, we show that the generated behaviors align with theoretical expectations for (A&amp;B). However, we also find that when gender is specified in the prompt, LLMs tend to reproduce societal gender stereotypes, associating male agents' behaviors with high ability and female agents' behaviors with high benevolence. To validate our approach, we conducted a user study on Prolific using a withinsubjects design. Participants perceived different levels of (A&amp;B) in the generated behaviors align with the intended instructions.
• LCoT-GV: Graph Attention Networks for Verifying Long Reasoning Chains in Large Language Models
  
  • Jaulmes Bérénice
  • Alam Mehwish
  , 2026. Large Reasoning Models produce Long Chains-of-Thought (LCoTs) which involve breaking down the problem into smaller reasoning steps before reaching the conclusion. However, these steps often contain contradictions, unsupported inferences, or irrelevant steps, even when the final answer is correct. We propose Long Chain-of-Thought Graph Verifier (LCoT-GV), a graph-based framework that represents LCoTs as reasoning graphs. Each node in the graph represents a reasoning step and the edges encode semantic and logical relations. A Graph Attention Network is then trained to predict chain-of-thought correctness from the reasoning graph. We construct a new graph-oriented verification dataset from multiple reasoning benchmarks for question answering in various domains. The results show that our method is competitive with the most similar approaches.
• Scale-Aware Feature-Level Encoding for Historical Document Stamp Detection
  
  • Chelali Mohamed
  • Cloppet Florence
  • Kurtz Camille
  • Bloch Isabelle
  • Foliard Daniel
  , 2026, 16851, pp.195-212. Stamps are generally used to authenticate, validate, and officially identify a document. Their analysis is particularly complex, especially for historical documents, which can be damaged. This article focuses on the particular task of stamp detection from old photographs. These stamps allow historians to trace the provenance and distribution of photographs within historical press agencies and, therefore, convey important information. Their varied, sometimes complex shapes, their non-standardized spatial location on the document, and the presence of textual information at different scales make their detection very challenging. In addition, the print quality of the stamp on the paper also has an impact, as some stamps may be faded or even completely invisible. To cope with the multi-scale nature of the features to be detected and with the potential degradation of scanned documents, we introduce a Scale-Aware Positional Encoding (SAPE) module designed for vision Transformer-based architectures that rely on multi-scale feature maps. Our module encodes the normalized logarithmic stride as a continuous variable using two complementary components: (1) an interpolation within a learnable table of scale anchors, and (2) a sinusoidal projection capturing fine-grained variations. The resulting embedding provides a smooth and adaptive modulation of multi-resolution features, leading to better spatial coherence across feature-pyramid levels. Experimental results demonstrate that SAPE improves the DeepSolo baseline and achieves competitive performance against representative detection approaches. These results confirm the effectiveness of SAPE in handling high scale variability and degradation in archival scans. The code is available at: https://github.com/mchelali/SA-DETR. (10.1007/978-3-032-36207-0_12)
  DOI : 10.1007/978-3-032-36207-0_12
• HELIOS: Lightweight real-time scheduling with energy readiness guarantees
  
  • Qiu Ruizhe
  • Robert Thomas
  • Tardieu Samuel
  • Pautet Laurent
  • Singhoff Frank
  Journal of Systems Architecture, Elsevier, 2026, 178, pp.1038856-1:1038856-10. Energy-harvesting embedded systems, like sensors and medical implants, must satisfy real-time constraints under strict energy limitations. Traditional schedulers, which prioritize timing over energy, often fail to produce valid schedules, particularly for non-preemptive task models. We introduce HELIOS, a lightweight, table-driven scheduler that treats timing and energy as joint first-class constraints while supporting limited-preemptive execution. HELIOS introduces energy readiness, ensuring jobs execute only when both temporal and energy conditions are met. The scheduling problem is formulated as an integer linear programming (ILP) problem, jointly optimizing task ordering, preemption points, and execution under energy constraints. By leveraging non-preemption and energy readiness, HELIOS maintains complexity proportional to job count. A key innovation is the use of worst-case energy footprints to guarantee safe execution via an energy readiness threshold. The table-driven runtime ensures compatibility with resource-constrained kernels. Experiments demonstrate that HELIOS achieves higher schedulability than priority-based baselines. (10.1016/j.sysarc.2026.103885)
  DOI : 10.1016/j.sysarc.2026.103885