In large organisations, identifying experts on a given topic is crucial in leveraging the internal knowledge spread across teams and departments. So-called enterprise expert retrieval systems automatically discover and structure employees' expertise based on the vast amount of heterogeneous data available about them and the work they perform. Evaluating these systems requires comprehensive ground truth expert annotations, which are hard to obtain. Therefore, the annotation process typically relies on automated recommendations of knowledge areas to validate. This case study provides an analysis of how these recommendations can impact the evaluation of expert finding systems. We demonstrate on a popular benchmark that system-validated annotations lead to overestimated performance of traditional term-based retrieval models and even invalidate comparisons with more recent neural methods. We also augment knowledge areas with synonyms to uncover a strong bias towards literal mentions of their constituent words. Finally, we propose constraints to the annotation process to prevent these biased evaluations, and show that this still allows annotation suggestions of high utility. These findings should inform benchmark creation or selection for expert finding, to guarantee meaningful comparison of methods.

Language-based AI systems are diffusing into society, bringing positive and negative impacts. Mitigating negative impacts depends on accurate impact assessments, drawn from an empirical evidence base that makes causal connections between AI usage and impacts. Interconnected post-deployment monitoring combines information about model integration and use, application use, and incidents and impacts. For example, inference time monitoring of chain-of-thought reasoning can be combined with long-term monitoring of sectoral AI diffusion, impacts and incidents. Drawing on information sharing mechanisms in other industries, we highlight example data sources and specific data points that governments could collect to inform AI risk management.

Multimodal aerial data are used to monitor natural systems, and machine learning can significantly accelerate the classification of landscape features within such imagery to benefit ecology and conservation. It remains under-explored, however, how these multiple modalities ought to be fused in a deep learning model. As a step towards filling this gap, we study three strategies (Early fusion, Late fusion, and Mixture of Experts) for fusing thermal, RGB, and LiDAR imagery using a dataset of spatially-aligned orthomosaics in these three modalities. In particular, we aim to map three ecologically-relevant biophysical landscape features in African savanna ecosystems: rhino middens, termite mounds, and water. The three fusion strategies differ in whether the modalities are fused early or late, and if late, whether the model learns fixed weights per modality for each class or generates weights for each class adaptively, based on the input. Overall, the three methods have similar macro-averaged performance with Late fusion achieving an AUC of 0.698, but their per-class performance varies strongly, with Early fusion achieving the best recall for middens and water and Mixture of Experts achieving the best recall for mounds.

Biological tree analysis serves as a pivotal tool in uncovering the evolutionary and differentiation relationships among organisms, genes, and cells. Its applications span diverse fields including phylogenetics, developmental biology, ecology, and medicine. Traditional tree inference methods, while foundational in early studies, face increasing limitations in processing the large-scale, complex datasets generated by modern high-throughput technologies. Recent advances in deep learning offer promising solutions, providing enhanced data processing and pattern recognition capabilities. However, challenges remain, particularly in accurately representing the inherently discrete and non-Euclidean nature of biological trees. In this review, we first outline the key biological priors fundamental to phylogenetic and differentiation tree analyses, facilitating a deeper interdisciplinary understanding between deep learning researchers and biologists. We then systematically examine the commonly used data formats and databases, serving as a comprehensive resource for model testing and development. We provide a critical analysis of traditional tree generation methods, exploring their underlying biological assumptions, technical characteristics, and limitations. Current developments in deep learning-based tree generation are reviewed, highlighting both recent advancements and existing challenges. Furthermore, we discuss the diverse applications of biological trees across various biological domains. Finally, we propose potential future directions and trends in leveraging deep learning for biological tree research, aiming to guide further exploration and innovation in this field.

Compositionality, the notion that the meaning of an expression is constructed from the meaning of its parts and syntactic rules, permits the infinite productivity of human language. For the first time, artificial language models (LMs) are able to match human performance in a number of compositional generalization tasks. However, much remains to be understood about the representational mechanisms underlying these abilities. We take a high-level geometric approach to this problem by relating the degree of compositionality in a dataset to the intrinsic dimensionality of its representations under an LM, a measure of feature complexity. We find not only that the degree of dataset compositionality is reflected in representations' intrinsic dimensionality, but that the relationship between compositionality and geometric complexity arises due to learned linguistic features over training. Finally, our analyses reveal a striking contrast between linear and nonlinear dimensionality, showing that they respectively encode formal and semantic aspects of linguistic composition.

Large language model (LLM)-based AI delegates are increasingly utilized to act on behalf of users, assisting them with a wide range of tasks through conversational interfaces. Despite their advantages, concerns arise regarding the potential risk of privacy leaks, particularly in scenarios involving social interactions. While existing research has focused on protecting privacy by limiting the access of AI delegates to sensitive user information, many social scenarios require disclosing private details to achieve desired outcomes, necessitating a balance between privacy protection and disclosure. To address this challenge, we conduct a pilot study to investigate user preferences for AI delegates across various social relations and task scenarios, and then propose a novel AI delegate system that enables privacy-conscious self-disclosure. Our user study demonstrates that the proposed AI delegate strategically protects privacy, pioneering its use in diverse and dynamic social interactions.

We present visual inertial lidar legged navigation system (VILENS), an odometry system for legged robots based on factor graphs. The key novelty is the tight fusion of four different sensor modalities to achieve reliable operation when the individual sensors would otherwise produce degenerate estimation. To minimize leg odometry drift, we extend the robot's state with a linear velocity bias term, which is estimated online. This bias is observable because of the tight fusion of this preintegrated velocity factor with vision, lidar, and inertial measurement unit (IMU) factors. Extensive experimental validation on different ANYmal quadruped robots is presented, for a total duration of 2 h and 1.8 km traveled. The experiments involved dynamic locomotion over loose rocks, slopes, and mud, which caused challenges such as slippage and terrain deformation. Perceptual challenges included dark and dusty underground caverns, and open and feature-deprived areas. We show an average improvement of 62% translational and 51% rotational errors compared to a state-of-the-art loosely coupled approach. To demonstrate its robustness, VILENS was also integrated with a perceptive controller and a local path planner.

Strictly speaking, many existing "subgradient" algorithms for training deep neural networks (e.g., [1, 2]) are not necessarily subgradient-based in any conventional sense due to the possible failure of equality-type subdifferential chain rules for the underlying problems. Consequently, it is often difficult to establish rigorous theoretical guarantees for these algorithms. Therefore, it is meaningful to study conditions under which equality-type subdifferential chain rules hold for nonsmooth functions that arise in applications. Apart from this perspective, subdifferential chain rules are also helpful in some other tasks, e.g., stationarity testing [3] and landscape analysis [4]. Unfortunately, the research along this direction is still quite limited. In the recent work [5], an efficiently-verifiable necessary and sufficient data qualification for the validity of subdifferential chain rules in training two-layer ReLU neural networks is developed. Motivated by this, it is very natural to ask if we can further extend these theories to other more sophisticated neural networks, such as those carrying some multilinear structures (instead of the linearity underlying the ReLU neural networks), which are prevalent in recommender systems studies; see, e.g., [6, 7, 8, 9, 10, 11, 12, 13, 14]. However, due to the multilinear structure involved, the situation becomes drastically different from the linear setting; e.g., the Jacobian of any multilinear mapping at the origin is always zero, which prevents the application of the well-known surjectivity condition for subdifferential chain rules (see, e.g., [15, Exercise 10.7]). As a consequence, to achieve the goal, new techniques have to be developed.

Questions involving commonsense reasoning about everyday situations often admit many $\textit{possible}$ or $\textit{plausible}$ answers. In contrast, multiple-choice question (MCQ) benchmarks for commonsense reasoning require a hard selection of a single correct answer, which, in principle, should represent the $\textit{most}$ plausible answer choice. On $250$ MCQ items sampled from two commonsense reasoning benchmarks, we collect $5,000$ independent plausibility judgments on answer choices. We find that for over 20% of the sampled MCQs, the answer choice rated most plausible does not match the benchmark gold answers; upon manual inspection, we confirm that this subset exhibits higher rates of problems like ambiguity or semantic mismatch between question and answer choices. Experiments with LLMs reveal low accuracy and high variation in performance on the subset, suggesting our plausibility criterion may be helpful in identifying more reliable benchmark items for commonsense evaluation.

Large Language Models (LLMs) have become pervasive due to their knowledge absorption and text-generation capabilities. Concurrently, the copyright issue for pretraining datasets has been a pressing concern, particularly when generation includes specific styles. Previous methods either focus on the defense of identical copyrighted outputs or find interpretability by individual tokens with computational burdens. However, the gap between them exists, where direct assessments of how dataset contributions impact LLM outputs are missing. Once the model providers ensure copyright protection for data holders, a more mature LLM community can be established. To address these limitations, we introduce CopyLens, a new framework to analyze how copyrighted datasets may influence LLM responses. Specifically, a two-stage approach is employed: First, based on the uniqueness of pretraining data in the embedding space, token representations are initially fused for potential copyrighted texts, followed by a lightweight LSTM-based network to analyze dataset contributions. With such a prior, a contrastive-learning-based non-copyright OOD detector is designed. Our framework can dynamically face different situations and bridge the gap between current copyright detection methods. Experiments show that CopyLens improves efficiency and accuracy by 15.2% over our proposed baseline, 58.7% over prompt engineering methods, and 0.21 AUC over OOD detection baselines.