Despite their growing adoption across domains, large language model (LLM)-powered agents face significant security risks from backdoor attacks during training and fine-tuning. These compromised agents can subsequently be manipulated to execute malicious operations when presented with specific triggers in their inputs or environments. To address this pressing risk, we present ReAgent, a novel defense against a range of backdoor attacks on LLM-based agents. Intuitively, backdoor attacks often result in inconsistencies among the user's instruction, the agent's planning, and its execution. Drawing on this insight, ReAgent employs a two-level approach to detect potential backdoors. At the execution level, ReAgent verifies consistency between the agent's thoughts and actions; at the planning level, ReAgent leverages the agent's capability to reconstruct the instruction based on its thought trajectory, checking for consistency between the reconstructed instruction and the user's instruction. Extensive evaluation demonstrates ReAgent's effectiveness against various backdoor attacks across tasks. For instance, ReAgent reduces the attack success rate by up to 90\% in database operation tasks, outperforming existing defenses by large margins. This work reveals the potential of utilizing compromised agents themselves to mitigate backdoor risks.

Personalized recommendation systems must adapt to user interactions across different domains. Traditional approaches like MLoRA apply a single adaptation per domain but lack flexibility in handling diverse user behaviors. To address this, we propose MoE-MLoRA, a mixture-of-experts framework where each expert is first trained independently to specialize in its domain before a gating network is trained to weight their contributions dynamically. We evaluate MoE-MLoRA across eight CTR models on Movielens and Taobao, showing that it improves performance in large-scale, dynamic datasets (+1.45 Weighed-AUC in Taobao-20) but offers limited benefits in structured datasets with low domain diversity and sparsity. Further analysis of the number of experts per domain reveals that larger ensembles do not always improve performance, indicating the need for model-aware tuning. Our findings highlight the potential of expert-based architectures for multi-domain recommendation systems, demonstrating that task-aware specialization and adaptive gating can enhance predictive accuracy in complex environments. The implementation and code are available in our GitHub repository.

-- Code review is a fundamental practice in software engineering, ensuring code quality, fostering collaboration, and reducing defects. While research has extensively examined var - ious aspects of this process, most studies assume that all code reviews follow a standardized evaluation workflow. However, our industrial partner, which uses Merge Requests (MRs) mechanism for code review, reports that this assumption does not always hold in practice. Many MRs serve alternative purposes beyond rigorous code evaluation. These MRs often bypass the standard review process, requiring minimal oversight. We refer to thes e cases as deviations, as they disrupt expected workflow patterns. For example, work - in - progress (WIP) MRs may be used as draft implementations without the intention of being review ed, MRs with huge changes are often created for code rebase, and library updates typically involve dependency version changes that require minimal or no review effort. We hypothesize that overlooking MR deviations can lead to biased analytics and reduced reliability of machine learning (ML) models used to explain the code review process. Our findings show that deviations occur in up to 37.02% of MRs across seven distinct categories. In addition, we develop a detection approach leveraging few - shot learning, achieving up to 91% accuracy in identifying these deviations. Furthermore, we examine the impact of removing MR deviations on ML models predicting code review completion time. Removing deviations significantly enhances model performance in 53.33% of cases, with improvements of up to 2.25 times. Our contributions include: (1) a clear definition and catego - rization of MR deviations, (2) a novel AI - based detection method leveraging few - shot learning, and (3) an empirical analysis of their exclusion impact on ML models explaining code review complet ion time. Our approach helps practitioners streamline review workflows, allocate reviewer effort more effectively, and ensure more reliable insights from MR analytics.

--Machine learning models are being increasingly used to automate decisions in almost every domain, and ensuring the performance of these models is crucial for ensuring high quality machine learning enabled services. Ensuring concept drift is detected early is thus of the highest importance. A lot of research on concept drift has focused on the supervised case that assumes the true labels of supervised tasks are available immediately after making predictions. Controlling for false positives while monitoring the performance of predictive models used to make inference from extremely large datasets periodically, where the true labels are not instantly available, becomes extremely challenging. We propose a flexible and efficient concept drift detection algorithm that uses classical statistical process control in a label-less setting to accurately detect concept drifts. We shown empirically that under computational constraints, our approach has better statistical power than previous known methods. Furthermore, we introduce a new drift detection framework to model the scenario of detecting drift (without labels) given prior detections, and show our how our drift detection algorithm can be incorporated effectively into this framework. We demonstrate promising performance via numerical simulations.

Speech technology systems struggle with many downstream tasks for child speech due to small training corpora and the difficulties that child speech pose. We apply a novel dataset, SpeechMaturity, to state-of-the-art transformer models to address a fundamental classification task: identifying child vocalizations. Unlike previous corpora, our dataset captures maximally ecologically-valid child vocalizations across an unprecedented sample, comprising children acquiring 25+ languages in the U.S., Bolivia, Vanuatu, Papua New Guinea, Solomon Islands, and France. The dataset contains 242,004 labeled vocalizations, magnitudes larger than previous work. Models were trained to distinguish between cry, laughter, mature (consonant+vowel), and immature speech (just consonant or vowel). Models trained on the dataset outperform state-of-the-art models trained on previous datasets, achieved classification accuracy comparable to humans, and were robust across rural and urban settings.

--Graph Neural Networks (GNNs) often struggle with heterophilic data, where connected nodes may have dissimilar labels, as they typically assume homophily and rely on local message passing. T o address this, we propose creating alternative graph structures by linking nodes with similar structural attributes (e.g., role-based or global), thereby fostering higher label homophily on these new graphs. We theoretically prove that GNN performance can be improved by utilizing graphs with fewer false positive edges (connections between nodes of different classes) and that considering multiple graph views increases the likelihood of finding such beneficial structures. Building on these insights, we introduce Structure-Guided GNN (SG-GNN), an architecture that processes the original graph alongside the newly created structural graphs, adaptively learning to weigh their contributions. Extensive experiments on various benchmark datasets, particularly those with heterophilic characteristics, demonstrate that our SG-GNN achieves state-of-the-art or highly competitive performance, highlighting the efficacy of exploiting structural information to guide GNNs. RAPH neural networks (GNNs) have demonstrated remarkable performance in processing graph-structured data by leveraging local neighborhood information [1], [2]. For node classification or regression, the graph structure is traditionally assumed to indicate which nodes should share similar representations or be treated similarly by a GNN. This holds true in many real-world scenarios; for example, temperature readings on a graph discretizing a geographical region often exhibit smoothness, where nearby sensors record similar temperatures [2]. M. Tenorio, S. Rey, and A. G. Marques are with the Department of Signal Theory and Communications, King Juan Carlos University, Madrid, Spain.

We propose sparseGeoHOPCA, a novel framework for sparse higher-order principal component analysis (SHOPCA) that introduces a geometric perspective to high-dimensional tensor decomposition. By unfolding the input tensor along each mode and reformulating the resulting subproblems as structured binary linear optimization problems, our method transforms the original nonconvex sparse objective into a tractable geometric form. This eliminates the need for explicit covariance estimation and iterative deflation, enabling significant gains in both computational efficiency and interpretability, particularly in high-dimensional and unbalanced data scenarios. We theoretically establish the equivalence between the geometric subproblems and the original SHOPCA formulation, and derive worst-case approximation error bounds based on classical PCA residuals, providing data-dependent performance guarantees. The proposed algorithm achieves a total computational complexity of $O\left(\sum_{n=1}^{N} (k_n^3 + J_n k_n^2)\right)$, which scales linearly with tensor size. Extensive experiments demonstrate that sparseGeoHOPCA accurately recovers sparse supports in synthetic settings, preserves classification performance under 10$\times$ compression, and achieves high-quality image reconstruction on ImageNet, highlighting its robustness and versatility.

Large language models (LLMs) are increasingly being used for complex research tasks such as literature review, idea generation, and scientific paper analysis, yet their ability to truly understand and process the intricate relationships within complex research papers, such as the logical links between claims and supporting evidence remains largely unexplored. In this study, we present CLAIM-BENCH, a comprehensive benchmark for evaluating LLMs' capabilities in scientific claim-evidence extraction and validation, a task that reflects deeper comprehension of scientific argumentation. We systematically compare three approaches which are inspired by divide and conquer approaches, across six diverse LLMs, highlighting model-specific strengths and weaknesses in scientific comprehension. Through evaluation involving over 300 claim-evidence pairs across multiple research domains, we reveal significant limitations in LLMs' ability to process complex scientific content. Our results demonstrate that closed-source models like GPT-4 and Claude consistently outperform open-source counterparts in precision and recall across claim-evidence identification tasks. Furthermore, strategically designed three-pass and one-by-one prompting approaches significantly improve LLMs' abilities to accurately link dispersed evidence with claims, although this comes at increased computational cost. CLAIM-BENCH sets a new standard for evaluating scientific comprehension in LLMs, offering both a diagnostic tool and a path forward for building systems capable of deeper, more reliable reasoning across full-length papers.

To meet the demands of content moderation, online platforms have resorted to automated systems. Newer forms of real-time engagement($\textit{e.g.}$, users commenting on live streams) on platforms like Twitch exert additional pressures on the latency expected of such moderation systems. Despite their prevalence, relatively little is known about the effectiveness of these systems. In this paper, we conduct an audit of Twitch's automated moderation tool ($\texttt{AutoMod}$) to investigate its effectiveness in flagging hateful content. For our audit, we create streaming accounts to act as siloed test beds, and interface with the live chat using Twitch's APIs to send over $107,000$ comments collated from $4$ datasets. We measure $\texttt{AutoMod}$'s accuracy in flagging blatantly hateful content containing misogyny, racism, ableism and homophobia. Our experiments reveal that a large fraction of hateful messages, up to $94\%$ on some datasets, $\textit{bypass moderation}$. Contextual addition of slurs to these messages results in $100\%$ removal, revealing $\texttt{AutoMod}$'s reliance on slurs as a moderation signal. We also find that contrary to Twitch's community guidelines, $\texttt{AutoMod}$ blocks up to $89.5\%$ of benign examples that use sensitive words in pedagogical or empowering contexts. Overall, our audit points to large gaps in $\texttt{AutoMod}$'s capabilities and underscores the importance for such systems to understand context effectively.

I present a novel approach to estimating the temporal and spatial patterns of violent conflict. I show how we can use highly temporally and spatially disaggregated data on conflict events in tandem with Gaussian processes to estimate temporospatial conflict trends. These trends can be studied to gain insight into conflict traps, diffusion and tempo-spatial conflict exposure in general; they can also be used to control for such phenomenons given other estimation tasks; lastly, the approach allow us to extrapolate the estimated tempo-spatial conflict patterns into future temporal units, thus facilitating powerful, stat-of-the-art, conflict forecasts. Importantly, these results are achieved via a relatively parsimonious framework using only one data source: past conflict patterns.