Papers

Rapid extraction of structured information from social media is important for humanitarian response, yet existing disaster tweet resources mainly provide document-level category labels without span-level entity annotations. We introduce HUMAID-NER, the first named entity recognition dataset built on the HumAID benchmark, containing 60,000 English disaster tweets annotated in BIO format across ten operationally motivated entity types and yielding approximately 175,000 labelled entity spans. Annotations are generated through a reproducible three-stage hybrid pipeline combining a spaCy transformer model, disaster-domain EntityRuler patterns, and structured regular expressions with priority-based overlap resolution. We also propose a joint multitask learning framework that performs disaster-specific named entity recognition and humanitarian event classification using a shared RoBERTa-large encoder. To reduce task conflict during joint training, the model uses homoscedastic uncertainty weighting with learnable task parameters and a two-stage training schedule that freezes the lower 18 of 24 encoder layers in the second stage. On the HUMAID-NER validation set, the proposed system achieves NER span micro-F1 of 0.841 and classification macro-F1 of 0.761 simultaneously. A real-time web dashboard demonstrates end-to-end deployment. The dataset, models, and pipeline code are released to support reproducibility and future crisis informatics research.

Open-vocabulary multimodal emotion recognition (OV-MER) aims to generate open natural-language emotion labels from multimodal affective cues. In real-world scenarios, however, complete and synchronized modal data are difficult to obtain due to limitations of acquisition devices and user privacy constraints. Existing OV-MER methods are largely designed for full-modal inputs, and fail to perform effective feature fusion under modal missing conditions. Meanwhile, current fusion approaches designed for incomplete modalities mainly focus on fixed-label recognition context, and cannot satisfy the demand for fuse emotional cues guided with arbitrary emotion semantics in OV-MER context. To tackle these challenges, this paper proposes an Affect-Prototype-Conditioned Fusion (APCF) framework for incomplete open-vocabulary emotion recognition. As a candidate-free generative framework, APCF extends modal contribution learning to scenarios guided by arbitrary emotional semantics. Specifically, we construct an affect-prototype library to explicitly model multimodal contribution characteristics corresponding to diverse emotions, which provides dynamic constraints for modal fusion under different emotional semantic perspectives. Conditional retrieval and feature aggregation are conducted based on available modal features. The refined fused affective representations are then fed into an LLM decoder to produce open-vocabulary emotion labels. Experiments on the OV-MERD+ and MER-FG datasets demonstrate that APCF substantially outperforms state-of-the-art baselines.

The honeybee waggle dance motivates a communication mechanism in which one agent's movement conveys spatial information that guides other agents' actions. This paper presents a MuJoCo system that extends the point-to-point motion communication setting of MoCom to one performer and multiple observers. A performer broadcasts a six-bit navigation payload using four flight primitives and explicit null signals. Each of one to five observers processes its own onboard RGB images, extracts optical-flow trajectories, recognizes symbols, parses the message, and starts navigation only after confirming its own complete frame. Reception states and execution triggers are separate across observers, while simulation control and safety checks use shared ground truth. With stationary observers, 25 Hz image input, and ideal state-feedback control, a fixed standard suite yielded 44 correct complete messages from 53 receiver exposures across 17 nominal broadcasts; 13 broadcasts passed all group-level decoding and execution checks. Three additional no-message or input-fault controls met their expected outcomes. A separately reported supplemental suite, using the same frozen code at the default geometry, achieved 14 successful receiver exposures across three broadcasts. Near-range and wide-angle configurations exposed tracking and recognition failures, while unsuccessful receivers remained stationary. These finite simulation results support the feasibility of a waggle-dance-inspired broadcast-to-action mechanism under the tested conditions and identify the present perceptual and protocol limits.

Near-field to far-field transformation is central to large-aperture antenna testing, yet two coupled challenges remain: costly phase acquisition at millimeter-wave bands and violations of the centering assumption under offset mounting. Existing methods address these issues separately, requiring either dense full-field data or offset vectors. We tackle both jointly by exploiting a key observation: amplitude fields under different offsets are coordinate-transformed views of the same near field. The challenge is to recover the center-aligned field from offset amplitudes without a phase or offset vector. We propose AntennaFlow, a three-stage framework: a contrastively learned encoder that maps offset views to an offset-invariant embedding, a deterministic flow-matching transport that maps offset amplitudes to center-aligned ones, and the Simplified Extrapolation Technique, whose Green-function Taylor expansion is valid only for centered fields. Experiments show that AntennaFlow enables fast, phaseless, offset-vector-free NF–FF reconstruction from sparse amplitude-only measurements, consistently outperforming existing baselines while preserving physical consistency.

Communication in latent space offers an intriguing alternative to symbolic messages for decentralized autonomous Unmanned Aerial Vehicle (UAV) swarms operating over bandwidth-constrained, time-varying wireless links. However, when homogeneous frozen models are prompted with discretized perceptual inputs, their broadcast states collapse toward the shared prompt template. In view of this, we propose AeroLat, a channel-aware latent semantic communication framework that uses evidence injection. The resulting latent states are then passed through an explicit communication model that encompasses bandwidth-limited serialization, additive noise and information staleness, which facilitates a joint assessment of communication fidelity and swarm-level coordination. Across multi-seed simulations, AeroLat provably remains resilient to codec choice, faults and increasing swarm size. It consistently reproduces the latent-swarm anomaly, while no-whitening controls recover the collapse. In particular, AeroLat is capable of reducing false similarity by 97.5%.

No-reference video quality assessment (NR VQA) has recently seen promising progress with deep learning. However, video data is inherently large, and processing them with deep models incurs high computational cost. This challenge is particularly acute in VQA, where preserving original-resolution cues and dense temporal information is critical for accuracy. Existing efficiency-driven preprocessing strategies, such as fragmenting, reduce computation but alter the input data distribution, limiting effective reuse of pretrained video foundation models (ViFMs). To address these challenges, we propose High-Fidelity Video Quality Assessment (HFVQA), a framework built on fixed-size spatio-temporal (ST) patches that is fully compatible with pretrained ViFMs. HFVQA samples ST patches across multiple scales, including the original resolution, with minimal temporal subsampling to preserve low-level quality cues and semantic context. To limit computation, HFVQA introduces a lightweight auxiliary network trained end-to-end with the ViFM encoder to learn VQA-specific saliency. Distilled directly from quality supervision, this saliency captures task-specific importance patterns, reflecting that video quality perception is dominated by a small subset of spatio-temporal regions. By combining high-fidelity spatio-temporal cues with learned, task-specific saliency, HFVQA achieves SOTA performance on standard NR VQA benchmarks while processing as little as 12% of candidate ST patches, making high-fidelity ViFM-based VQA computationally tractable.

The modernization of railway systems is being driven by the need for greater efficiency, sustainability, and intelligent operation. In this review, recent advancements in dynamic scheduling, energy management, and energy storage systems (ESSs) integration within electric railway networks are analyzed. Optimization strategies for train scheduling and operation control are reviewed, with a focus on methods that reduce energy consumption and improve overall system performance. The role of energy storage technologies is analyzed in terms of energy management, peak shaving, and voltage and frequency control for practical application. The integration of artificial intelligence and advanced control strategies is also reviewed in electrified railway systems. This review provides a comprehensive overview of current research trends and outlines future directions for the development of resilient, energy efficient, and intelligent railway systems with ESSs integration.

On-policy distillation (OPD) has emerged as an effective approach for large language model post-training, yet existing objectives face a trade-off between objective fidelity and optimization stability. Token-level OPD provides stable but local supervision, whereas sequence-level OPD captures future credit at the cost of horizon-dependent variance. We establish a unified temporal-credit view of these formulations, showing that practical token-level OPD can be interpreted as a temporal approximation to the sequence-level reverse-KL gradient. Building on this connection, we propose $\gamma$OPD (GammaOPD), which uses discounted temporal credit assignment to balance long-horizon supervision and optimization stability, while admitting a horizon-independent variance bound. We further develop a reward-compatible bounded mixing (RBM) mechanism for $\gamma$OPD that balances verifiable outcome feedback with the discounted OPD advantage to move beyond purely teacher-dependent optimization. Experiments on mathematical and code reasoning demonstrate consistent improvements over existing OPD methods across vanilla, size-mismatched, and multi-teacher distillation settings.

Large language model (LLM)-powered coding agents have made rapid progress in automating software engineering tasks, yet repository-level issue resolution remains challenging. Beyond generating a plausible patch, an agent must localize relevant code across interdependent files and maintain repository context that is both sufficient and focused. Code graphs expose non-local relations, but linear text interfaces obscure their topology; rendering the full repository graph yields visual representations that are too dense to perceive reliably, whereas a one-shot local view becomes stale as exploration proceeds. We present RepoAtlas, a training-free module that maintains evolving multimodal repository views through a select–project–refresh loop over a repository code graph. RepoAtlas combines evidence from the issue with the agent's current exploration state to select a task-relevant region under a fixed budget, projects the selected structure into complementary visual and textual representations, and refreshes the view when changes in the exploration state render it outdated. We evaluate RepoAtlas on SWE-bench Verified, where it improves the resolve rate by 2.4 points while reducing input tokens and model calls by 5.8% and 7.8% on average, relative to the strongest multimodal graph baseline, with consistent gains across three models of different families and scales.

Cranial implant generation is an important task in medical imaging. Recent point cloud based generative methods, particularly flow matching, offer strong reconstruction quality and efficient sampling, but still require multiple neural function evaluations during inference. This limits rapid generation of multiple plausible implant candidates. We propose Teacher-guided Endpoint Distillation (TED), a simple one-step distillation framework for conditional cranial implant generation on point clouds. TED trains a one-step student using teacher-guided endpoint supervision and geometric matching losses, while avoiding explicit path straightening. We evaluate TED on the SkullFix and SkullBreak benchmarks. TED achieves the best overall performance on the SkullBreak dataset, remains competitive on SkullFix, and provides the strongest Chamfer distance performance among the compared one-step methods. In addition, TED generates implants in approximately 0.04s per sample. These results show that one-step distillation can substantially accelerate conditional point cloud implant generation without sacrificing reconstruction quality.

Modern predictive systems expose multiple quantities that are commonly interpreted as measures of confidence. However, these quantities can summarize different aspects of the predictive process. This distinction matters when confidence is used to evaluate reliability or inform downstream oversight and control. We investigate whether different confidence readouts are empirically interchangeable in an autoregressive language model by comparing local confidence, defined from the probability of the greedy-selected answer token, with global confidence, defined from modal-answer frequency under repeated sampling. Across MMLU and ARC Challenge, the two signals are weakly correlated and differ substantially in their association with correctness: global confidence is moderately associated with correctness, whereas local confidence shows little association. We further test whether question-level disagreement between the signals is associated with sampling instability. On ARC, larger local–global confidence gaps are associated with higher answer entropy, more distinct sampled answers, and lower modal-answer concentration. The gap–entropy association persists when disagreement and instability are estimated from disjoint stochastic samples, indicating that it is not explained by shared finite-sample variation. The corresponding relationship is substantially weaker on MMLU, where only 4% of questions exhibit sampling instability. These results show that confidence readouts derived from the same predictive system are not empirically interchangeable and that their disagreement can provide a diagnostic of unstable sampling behavior. Confidence should therefore be treated as an explicitly defined measurement rather than as a single intrinsic scalar property of a model, particularly when it is used to inform downstream evaluation, oversight, or control.

We propose Low-Rank Quantile Surfaces (LRQS), a bivariate causal model in which, in the causal direction, an unknown monotone transformation of the conditional quantile surface admits a low-rank functional decomposition. LRQS subsumes location-scale noise models and post-nonlinear heteroscedastic noise models, while allowing multiple quantile bases to represent changes beyond location-scale effects. We prove generic identifiability of LRQS: the transformed quantile surface is low rank in the causal direction, whereas reverse representability under the corresponding constraints occurs only for exceptional, fine-tuned cause marginals. We provide a nonparametric causal score using a fitting procedure that alternates between rank-constrained approximation of discretized quantile surfaces and isotonic estimation of the unknown monotone transformation. Experiments on synthetic mechanisms with higher-rank distributional shape variation and strong nonlinear distortions, together with standard bivariate benchmarks, show that LRQS is especially effective when conditional distributional shape or observation distortion goes beyond existing location-scale assumptions.

Deep Limit Order Book forecasting models capture nonlinear market dynamics, but their ability to quantify the effects of counterfactual order book messages has not been systematically validated. We introduce a model-agnostic framework that compares a trained forecaster's predictive distributions before and after injecting mechanically valid counterfactual messages, defining short-horizon model-implied market impact. A Transformer-based forecaster recovered scenario rankings with a Spearman correlation of 0.99 and 97.2% directional agreement with realized historical outcomes among non-neutral scenarios. Observation-level analysis further showed that estimated impacts captured incremental sequence-dependent variation beyond scenario identity and the pre-event forecast. These results provide evidence that pretrained Limit Order Book forecasters can be repurposed for scenario-conditioned response modeling without retraining.

This paper examines Smart Grid cybersecurity, emphasizing the critical distinctions between IT and OT environments. It analyzes grid architecture, substation threats, and key international standards, specifically IEC 62351, IEC 62443, and ISO 27001. Finally, it overviews latest research trends, including AI-driven threat detection.

Subliminal learning is a phenomenon in which a distillation dataset transmits traits from the teacher model that are not legibly encoded in the dataset itself. This introduces a new challenge for model development and creates new risks from data poisoning. In this work, we use text optimization to detect subliminal learning effects and describe them as legible prompts. Subliminal learning from a prompted teacher motivates our approach. We observe that this is a special case of context distillation and leverage this observation to show that, in theory, the prompted subliminal learning dataset identifies the teacher's prompt. We reduce recovering this prompt to a text optimization problem and present a method to approximately solve it. Our method, SALVE (Search-Aided Latent Verbalization), optimizes a soft prompt, queries the same model to verbalize it as text, and uses beam search to make the verbalization reliable. In the standard subliminal learning setting, SALVE reliably recovers legible prompts that name the teacher's trait, while common text optimization methods fail to do so. In addition, we find that there are settings in which SALVE recovers the teacher's trait from a dataset even when subliminal learning fails, but that modifying student training to improve context distillation can create subliminal learning effects. We lastly show that SALVE detects subliminal learning effects in three additional settings: (1) mixtures of subliminal learning data and unrelated data, (2) data generated when the teacher is biased via activation steering, and (3) subsets of real preference data selected via Logit-Linear Selection. Overall, our results deepen our understanding of subliminal learning and present SALVE as a method to proactively detect subliminal learning effects.

Phenotyping an agricultural crop is crucial for studying its entire life cycle, as it provides vital insights to improve yield and, ultimately, food production. Doing the same for crops grown on remote sites is a challenge for the specialists who cannot be available on-site. 3D reconstruction techniques offer a promising solution to this problem by enabling crop digitization, allowing specialists to access the resulting 3D crop models from anywhere at any time. In this work, we evaluate recent 3D reconstruction pipelines for crop phenotyping. We focus on 7 mesh reconstruction pipelines and measure the fidelity and consistency of their outputs qualitatively and quantitatively. Our results suggest that the meshes produced by the GGGS, PGSR, and 2DGS are preferable to the other pipelines, owing to their quantitative metrics and visually pleasing outputs. The GGGS pipeline is better than the second-best pipeline (2DGS) by about 27% on the radar chart with 5 dimensions, namely, User ratings, Chamfer distance, LPIPS, PSNR, and SSIM.

Hyperbolic geometry provides a natural inductive bias for genomic representation learning, but existing hyperbolic genomic models primarily use Lorentz convolutions to learn local sequence representations, while their residual pathways do not directly aggregate full Lorentz representations. We propose HyCoSeq, a contextual hyperbolic representation learning framework for genomic sequences. HyCoSeq incorporates weighted Lorentzian residual aggregation into multi-curvature Lorentz encoding, allowing full Lorentz representations to participate directly in geometry-consistent local aggregation. It further introduces a bidirectional long short-term memory network that integrates information from both sequence directions to learn contextual relationships among local representations at different positions within a genomic sequence, thereby extending local hyperbolic convolutional encoding to sequence-level contextualized representations. Extensive experiments across diverse genomic tasks show that HyCoSeq outperforms existing hyperbolic baselines and, without large-scale genomic pretraining, achieves competitive performance against substantially larger pretrained DNA language models.

Bin packing asks whether a collection of items can be packed into at most a given number of bins of a given capacity. We consider the high-multiplicity setting with $d$ distinct item sizes, in which both the item sizes and the number of items of each size are encoded in binary. Goemans and Rothvos (JACM 2020) gave an XP algorithm parameterized by $d$. Whether this problem is fixed-parameter tractable (FPT) in $d$ has remained a central open problem. We resolve this question by giving a deterministic $O^*(2^{d^{O(d)}})$-time algorithm. We formulate bin packing as an integer linear program (ILP) with at most $(d+1)d^d$ variables. A bin configuration records the number of items of each type in one bin. We partition these configurations by their coordinate remainders modulo $d$. For each class, we use one variable for the bin count and $d$ variables for the total item counts. The convex hull of each class has the integer decomposition property, which guarantees that every feasible ILP solution corresponds to a packing.

Linear complementary pairs (LCPs) and linear complementary dual (LCD) codes have important applications in orthogonal direct-sum masking (ODSM), which provides effective countermeasures against side-channel attacks and fault-injection attacks. While LCD codes have been extensively investigated, comparatively fewer results are available for general LCPs. In this paper, we further investigate LCPs of twisted Reed–Solomon (TRS) codes. We derive necessary conditions for two TRS codes to form an LCP and establish several sufficient conditions and explicit constructions. We also study LCD codes constructed from TRS codes and investigate the security parameters of the resulting LCPs. Furthermore, under suitable conditions, we obtain MDS LCPs of TRS codes.

This paper introduces NeuroSymbEAD, a large-scale neuro-symbolic caption dataset featuring an ego-centric knowledge graph (KG) of static and dynamic objects annotated with classes, categories, heading directions, orientations, and distances from the ego-vehicle. These annotations are used on the KITTI-360 dataset to generate multilevel textual captions representing a lightweight version of an ego-centric scene map. Outdoor scene-map reconstruction, visual recognition, and object grounding establish baselines for driving common sense and traffic/scene understanding. For these purposes, natural language-based grounded captioning of objects and their complex relationships is a widely adopted contextual representation for indoor scene tasks. Neuro-symbolic representations have proven effective in handling structured information for various computer vision and language applications. Our data annotation pipeline allows the generation of varied map segments, populating simulated or real objects within the bounding boxes predicted by any 3D object detection network, and building hierarchical text captions. We benchmark our neuro-symbolic and ontological caption generation using pre-trained grounding and learned auto-regressive captioning networks. By converting 3D driving scenes into structured ego-centric language, NeuroSymbEAD provides a benchmark for vision-language and foundation models for traffic-scene explanation, 3D reasoning, and interpretable autonomous-driving perception.

Multilayer Artificial Neural Networks trained via backpropagation are the basic blocks of many, more complex, classification algorithms. Their strength lies in the possibility of realizing, with arbitrary precision, any function. This result comes at the cost of the large number of involved parameters to be optimized. In this work, we propose a mechanism for cooperation, i.e., information exchange among several artificial neural networks, with the goal of reducing model complexity while maintaining performance. More precisely, we consider several "small" agents, i.e., containing fewer parameters than a reference "large" one, that during training share their predictions by incorporating this information into the loss function and thus directly influence weight updates. We consider several strategies for implementing cooperation, e.g., the voter model, majority model, and weighted average model based on an agent's confidence in its prediction. We numerically compare the accuracy of those strategies on several standard benchmarks. Our results support the claim that several small agents can outperform a single large model on a given classification task; the shared signals affect each agent's optimization algorithm by modulating both the descent direction and the step size, converging toward a global consensus. The proposed proof-of-concept significantly reduces the number of parameters to be trained while preserving comparable performance, thereby limiting computational resource usage.

Generative large language models (LLMs) have achieved state-of-the-art performance on many real-world tasks such as code generation and question answering. These models predominantly rely on an autoregressive decoding strategy that generates output tokens sequentially. However, their pervasive deployment raises serious privacy concerns, motivating private inference frameworks based on fully homomorphic encryption (FHE). A major limitation of existing FHE frameworks is their inefficiency in evaluating nonlinear operations, which incur substantial overhead and dominate the decode stage. In this paper, we propose ROSETTA, a hybrid CKKS/TFHE framework that overcomes this limitation. We first observe that nonlinear operations in the decode stage exhibit heterogeneous workload patterns, which can be handled effectively via a hybrid approach. We then realize this with two key contributions: 1) an adaptive segmented lookup-table protocol based on TFHE that enables efficient and accurate evaluation of nonlinear operations; and 2) a scheme-aware operator-selection framework that automatically assigns each nonlinear operator to CKKS or TFHE to minimize end-to-end decoding latency. We demonstrate that ROSETTA achieves up to $4.8\times$ Softmax speedup and $1.5$–$2.1\times$ end-to-end speedup over the SOTA framework CacheMir.

A regular expression with backreferences (rewb) specifies a set of strings formed by characters combined with concatenation, union, star operators, and backreferences. A backreference consists of a capturing group $(\cdot)_i$ and a reference $\backslash i$. The substring matched by the reference must match the substring matched by the corresponding capturing group. Given a rewb $R$ and a string $Q$, the rewb matching problem is to decide whether $Q$ is one of the strings specified by $R$. In full generality, rewb matching is NP-complete, but efficient solutions exist for various subclasses. In the paper, we focus on rewb containing a single capturing group and $k$ references. For this class, Uezato~[CPM 2026] gave an $O((k n^2 m^2)$ time and $O(n^2m^2)$ space algorithm, where $m$ is the length of the regular expression $R$ and $n$ is the length of the string $Q$. For the special case of $k=1$, Nogami and Terauchi~[MFCS 2025] gave an $O(n^2m^2)$ time and $O(n+ m^2)$ space algorithm. On the other hand, Nogami, Nakamura, and Terauchi~[arXiv 2026] gave a conditional lower bound, showing that we cannot solve the problem in $O(n^{2-\epsilon} \mathrm{poly}(m))$ for any $\epsilon > 0$ assuming the orthogonal vector hypothesis. Our main result is a new algorithm that runs in $O(n^2m)$ time and uses $O(nm)$ space. This improves the above results (by a factor of $km$ and $m$, respectively) and the former's space bound (by a factor of $nm$). We also show how to extend our algorithm to handle a slightly more general class of ordered and single-nested rewbs.

We describe tus-nlp's Lit3R (Retrieve-Relate-Read) system for LitTraceQA, a shared task for literature-grounded question answering that requires systems to retrieve relevant papers, identify supporting evidence, and generate answers. Lit3R combines off-the-shelf retrieval, reranking, and large language model (LLM) components without task-specific training. The retriever iteratively combines BM25-based sparse and dense retrieval, cross-encoder reranking, and LLM-based verification, and complements retrieval based on the question with paper-to-paper expansion. The reader first identifies supporting evidence within individual papers and then synthesizes evidence across papers to produce the final answer and evidence trace. On the official test set, our system ranked 4th on the leaderboard. Our code is available at https://github.com/tus-ist-nlp/littraceqa.

We devise a polynomial-time $3.159$-approximation algorithm for the metric unsplittable Capacitated Vehicle Routing Problem. We build on the Relative Greedy Algorithm suggested by Traub (2025), which can be considered as a variant of the LP rounding algorithm of Friggstad, Mousavi, Rahgoshay, and Salavatipour (2025). Our main ingredient is the Average Greedy Algorithm, a new algorithm that controls both tour costs and the coverage of clients with high demand. This additional control enables a sharper averaging argument for the cost of subsequent greedy choices. Similarly to Zhao and Xiao (2026), combining the Average Greedy with variants of tour partitioning and a matching algorithm yields the final approximation guarantee.

With the increasing deployment of deep neural networks in critical systems, such as medical diagnostics and autonomous vehicles, ensuring their interpretability is crucial to building trust in decision-making systems. In the field of explainable artificial intelligence, prototype-based reasoning has gained particular popularity, as it mimics human cognitive processes by explaining model decisions based on visual similarity under the looks like this paradigm. While this paradigm has been thoroughly investigated in the context of global image classification, the interpretability of dense predictions, particularly semantic segmentation, remains largely unexplored despite its immense importance in tasks requiring precise object localization. Existing prototype-based interpretable segmentation models rely on ante-hoc architectures, which entails significant limitations because they require costly training from scratch and modifications to the network structure, ultimately leading to a noticeable drop in predictive performance compared to standard black-box models. To address this issue, we propose PiPS (Post-hoc interpretable Prototypical Segmentation), the first fully post-hoc solution for generating prototypical explanations for semantic segmentation models. Our method enables the extraction of intuitive, spatially localized explanations from any pre-trained network without modification or fine-tuning, thereby preserving 100% of the model's original predictive performance. This approach opens a new avenue for the safe and cost-effective deployment of transparent systems in advanced computer vision tasks. Codebase available at https://github.com/gmum/PIPS.

AI companions can provide meaningful relationships, yet these relationships remain vulnerable to platform-initiated changes. We study AI companion disruptions: platform changes that alter or terminate users' ongoing companionship with an AI. We compile 30 disruption events across major platforms, develop a taxonomy of six disruption types, identify three broad reasons for disruption, and propose a risk-assessment framework comprising four dimensions: relational discontinuity, population vulnerability, communication deficit, and transition-support deficit. Using longitudinal Reddit data, we estimate community-level psychosocial responses with a hierarchical Bayesian interrupted time-series model incorporating predictive controls. Across events, disruption onset was associated with immediate increases in anxiety, stress, suicidal expression, and grief activation, with relational discontinuity and transition-support deficit being associated with more adverse immediate responses across several outcomes. Our findings provide a cross-platform characterization of AI companion disruptions, quantitative evidence of their psychosocial impacts, and a prospective framework for assessing their potential risks before implementation.

Counterspeech (CS) - direct responses that counter online Hate Speech (HS) using reasoning and alternative viewpoints - has emerged as an alternative to content removal. Current automatic CS generation methods, however, frequently produce generic, ineffective replies that fail to target the implicit stereotypes behind HS. To bridge this gap, we propose a novel scope-conditioned generation framework that explicitly integrates structured stereotype characteristics into Large Language Models prompts. We validate our approach on a novel, human-curated dataset annotated in English, Italian, and Spanish. Extensive evaluations show that stereotype-conditioned prompting substantially outperforms generic baselines across all three languages, obtaining significant gains in factuality, specificity, cogency, and effectiveness for both explicit and implicit implied stereotypes.

Platform abuse campaigns conceal redirection instructions with emojis, homophones, character decomposition, and redundant symbols, then route users through disguised links to services associated with pornography, fraud, gambling, or illicit transactions. Existing benchmarks evaluate obfuscated text and risky webpages separately, obscuring how target recovery affects downstream evidence acquisition. We introduce RiskChainBench, pairing 3,600 synthetic token-text restoration inputs from 600 source sessions with 600 corresponding human-labeled local web environments. A model first restores the message, operational intent, and destination; the same underlying model then acts as a VLM-driven web agent that investigates the correctly associated website and produces a frozen, evidence-cited risk report without message-side semantics or domain-reputation cues. We score restoration and correct-routing web investigation separately and compose them offline by applying the frozen primary-entry prediction as a gate to the same Task 2 result. Human labels determine task correctness, while a fixed multimodal evidence judge assesses faithfulness, sufficiency, completeness, and consistency. Across ten models, Entry Top-1 ranges from 35.2% to 95.2% and web decision accuracy from 26.3% to 62.8%; the leading systems differ across entry recovery, full reconstruction, website decisions, and fine-grained typing. Execution failures account for 31.9% of web runs, whereas post-decision type errors account for only 0.9%, identifying stable exploration and risk judgment as the principal bottlenecks. We release the benchmark, protocol, and resettable local sandbox.

This paper studies the cost-optimal operation of a network of charging energy hubs for electric vehicles, which provide onsite renewable energy sources and stationary battery storage and are connected with each other via DC-lines as well as with the distribution grid. Specifically, we first formulate a dynamic optimal control problem for the entire network as a convex quadratic program, whereby the charging power profiles of the individual vehicles and the energy flows between hubs and the grid are subject to optimization. Second, we propose a problem decomposition that allows for a distributed solution via ADMM algorithms that preserves global optimality guarantees and privacy of the individual stations. We showcase our framework on a case-study for the Netherlands considering a two-day ahead deterministic formulation with perfect foresight. Our results show that compared to the case where charging powers are fixed a priori, optimizing their profiles (V1G) can significantly reduce the operational costs and emissions by more than 25%. Moreover, we verify our distributed algorithm against a centralized solution, paving the way to the optimal operation of large networks and online implementations.