Online platforms usually suffer from user experience degradation due to near-duplicate items with similar visuals and text. While Multimodal Large Language Models (MLLMs) enable multimodal embedding, existing methods treat representations as black boxes, ignoring structural relationships (e.g., primary vs. auxiliary elements), leading to local structural collapse problem. To address this, inspired by Feature Integration Theory (FIT), we propose FITRep, the first attention-guided, white-box item representation framework for fine-grained item deduplication. FITRep consists of: (1) Concept Hierarchical Information Extraction (CHIE), using MLLMs to extract hierarchical semantic concepts; (2) Structure-Preserving Dimensionality Reduction (SPDR), an adaptive UMAP-based method for efficient information compression; and (3) FAISS-Based Clustering (FBC), a FAISS-based clustering that assigns each item a unique cluster id using FAISS. Deployed on Meituan's advertising system, FITRep achieves +3.60% CTR and +4.25% CPM gains in online A/B tests, demonstrating both effectiveness and real-world impact.

I NPUT S PECIFIC N EURAL N ETWORKS A P REPRINT Asghar Jadoon The University of Texas at Austin Austin TX, USA D. Thomas Seidl Sandia National Laboratories Albuquerque NM, USA Reese E. Jones Sandia National Laboratories Livermore CA, USA Jan Fuhg The University of Texas at Austin Austin TX, USA February 2025 Abstract Neural networks have emerged as powerful tools for mapping between inputs and outputs. However, their black-box nature limits the ability to encode or impose specific structural relationships between inputs and outputs. While various studies have introduced architectures that ensure the network's output adheres to a particular form in relation to certain inputs, the majority of these approaches impose constraints on only a single set of inputs, leaving others unconstrained. This paper introduces a novel neural network architecture, termed the Input Specific Neural Network (ISNN), which extends this concept by allowing scalar-valued outputs to be subject to multiple constraints. Specifically, the ISNN can enforce convexity in some inputs, non-decreasing monotonicity combined with convexity with respect to others, and simple non-decreasing monotonicity or arbitrary relationships with additional inputs. To the best of our knowledge, this is the first work that proposes a framework that simultaneously comprehensively imposes all these constraints. The paper presents two distinct ISNN architectures, along with equations for the first and second derivatives of the output with respect to the inputs. These networks are broadly applicable. In this work, we restrict their usage to solving problems in computational mechanics. In particular, we show how they can be effectively applied to fitting data-driven constitutive models. We remark, that due to their increased ability to implicitly model constraints, we can show that ISNNs require fewer inputs than existing input convex neural networks when modeling polyconvex hyperelastic functions. We then embed our trained data-driven constitutive laws into a finite element solver where significant time savings can be achieved by using explicit manual differentiation using the derived equations as opposed to automatic differentiation. Manual differentiation also enables seamless employment of trained ISNNs in commercial solvers where automatic differentiation may not be possible. We also show how ISNNs can be used to learn structural relationships between inputs and outputs via a binary gating mechanism.

Given that substantial amounts of domain-specific knowledge are stored in structured formats, such as web data organized through HTML, Large Language Models (LLMs) are expected to fully comprehend this structured information to broaden their applications in various real-world downstream tasks. Current approaches for applying LLMs to structured data fall into two main categories: serialization-based and operation-based methods. Both approaches, whether relying on serialization or using SQL-like operations as an intermediary, encounter difficulties in fully capturing structural relationships and effectively handling sparse data. To address these unique characteristics of structured data, we propose HyperG, a hypergraph-based generation framework aimed at enhancing LLMs' ability to process structured knowledge. Specifically, HyperG first augment sparse data with contextual information, leveraging the generative power of LLMs, and incorporate a prompt-attentive hypergraph learning (PHL) network to encode both the augmented information and the intricate structural relationships within the data. To validate the effectiveness and generalization of HyperG, we conduct extensive experiments across two different downstream tasks requiring structured knowledge.

Medication Recommendation (MR) is a promising research topic which booms diverse applications in the healthcare and clinical domains. However, existing methods mainly rely on sequential modeling and static graphs for representation learning, which ignore the dynamic correlations in diverse medical events of a patient's temporal visits, leading to insufficient global structural exploration on nodes. Additionally, mitigating drug-drug interactions (DDIs) is another issue determining the utility of the MR systems. To address the challenges mentioned above, this paper proposes a novel MR method with the integration of dynamic networks and multi-view drug representations (DNMDR). Specifically, weighted snapshot sequences for dynamic heterogeneous networks are constructed based on discrete visits in temporal EHRs, and all the dynamic networks are jointly trained to gain both structural correlations in diverse medical events and temporal dependency in historical health conditions, for achieving comprehensive patient representations with both semantic features and structural relationships. Moreover, combining the drug co-occurrences and adverse drug-drug interactions (DDIs) in internal view of drug molecule structure and interactive view of drug pairs, the safe drug representations are available to obtain high-quality medication combination recommendation. Finally, extensive experiments on real world datasets are conducted for performance evaluation, and the experimental results demonstrate that the proposed DNMDR method outperforms the state-of-the-art baseline models with a large margin on various metrics such as PRAUC, Jaccard, DDI rates and so on.

Large language models record impressive performance on many natural language processing tasks. However, their knowledge capacity is limited to the pretraining corpus. Retrieval augmentation offers an effective solution by retrieving context from external knowledge sources to complement the language model. However, existing retrieval augmentation techniques ignore the structural relationships between these documents. Furthermore, retrieval models are not explored much in scientific tasks, especially in regard to the faithfulness of retrieved documents. In this paper, we propose a novel structure-aware retrieval augmented language model that accommodates document structure during retrieval augmentation. We create a heterogeneous document graph capturing multiple types of relationships (e.g., citation, co-authorship, etc.) that connect documents from more than 15 scientific disciplines (e.g., Physics, Medicine, Chemistry, etc.). We train a graph neural network on the curated document graph to act as a structural encoder for the corresponding passages retrieved during the model pretraining. Particularly, along with text embeddings of the retrieved passages, we obtain structural embeddings of the documents (passages) and fuse them together before feeding them to the language model. We evaluate our model extensively on various scientific benchmarks that include science question-answering and scientific document classification tasks. Experimental results demonstrate that structure-aware retrieval improves retrieving more coherent, faithful and contextually relevant passages, while showing a comparable performance in the overall accuracy.

We study the piecewise stationary combinatorial semi-bandit problem with causally related rewards. In our nonstationary environment, variations in the base arms' distributions, causal relationships between rewards, or both, change the reward generation process. In such an environment, an optimal decision-maker must follow both sources of change and adapt accordingly. The problem becomes aggravated in the combinatorial semi-bandit setting, where the decision-maker only observes the outcome of the selected bundle of arms. The core of our proposed policy is the Upper Confidence Bound (UCB) algorithm. We assume the agent relies on an adaptive approach to overcome the challenge. More specifically, it employs a change-point detector based on the Generalized Likelihood Ratio (GLR) test. Besides, we introduce the notion of group restart as a new alternative restarting strategy in the decision making process in structured environments. Finally, our algorithm integrates a mechanism to trace the variations of the underlying graph structure, which captures the causal relationships between the rewards in the bandit setting. Theoretically, we establish a regret upper bound that reflects the effects of the number of structural- and distribution changes on the performance. The outcome of our numerical experiments in real-world scenarios exhibits applicability and superior performance of our proposal compared to the state-of-the-art benchmarks.

We describe an unconventional line of attack in our quest to teach machines how to rap battle by improvising hip hop lyrics on the fly, in which a novel recursive bilingual neural network, TRAAM, implicitly learns soft, context-dependent generalizations over the structural relationships between associated parts of challenge and response raps, while avoiding the exponential complexity costs that symbolic models would require. TRAAM learns feature vectors simultaneously using context from both the challenge and the response, such that challenge-response association patterns with similar structure tend to have similar vectors. Improvisation is modeled as a quasi-translation learning problem, where TRAAM is trained to improvise fluent and rhyming responses to challenge lyrics. The soft structural relationships learned by our TRAAM model are used to improve the probabilistic responses generated by our improvisational response component.

Despite substantial progress in applying neural networks (NN) to a wide variety of areas, they still largely suffer from a lack of transparency and interpretability. While recent developments in explainable artificial intelligence attempt to bridge this gap (e.g., by visualizing the correlation between input pixels and final outputs), these approaches are limited to explaining low-level relationships, and crucially, do not provide insights on error correction. In this work, we propose a framework (VRX) to interpret classification NNs with intuitive structural visual concepts. Given a trained classification model, the proposed VRX extracts relevant class-specific visual concepts and organizes them using structural concept graphs (SCG) based on pairwise concept relationships. By means of knowledge distillation, we show VRX can take a step towards mimicking the reasoning process of NNs and provide logical, concept-level explanations for final model decisions. With extensive experiments, we empirically show VRX can meaningfully answer "why" and "why not" questions about the prediction, providing easy-to-understand insights about the reasoning process. We also show that these insights can potentially provide guidance on improving NN's performance.

This paper presents a model architecture for encoding the representations of part-whole hierarchies in images in form of a graph. The idea is to divide the image into patches of different levels and then treat all of these patches as nodes for a fully connected graph. A dynamic feature extraction module is used to extract feature representations from these patches in each graph iteration. This enables us to learn a rich graph representation of the image that encompasses the inherent part-whole hierarchical information. Utilizing proper self-supervised training techniques, such a model can be trained as a general purpose vision encoder model which can then be used for various vision related downstream tasks (e.g., Image Classification, Object Detection, Image Captioning, etc.).

CausalNex is a Python library that allows data scientists and domain experts to co-develop models that go beyond correlation and consider causal relationships. 'CasualNex' provides a practical'what if' library which is deployed to test scenarios using Bayesian Networks (BNs). 'CasualNex' prepares practitioners to understand structural relationships from data and helps in the verification for accuracy of the relationships between different data sets. Apart from practitioners understanding the structural relationship from data, it also enables domain experts to fit conditional probability distributions and study the effect of potential interventions. 'CasualNex' helps to simplify the following steps: CausalNex is a Python package.