Searcharxiv⌕ Search

arXiv subjects

Donald Loveland

Publications and source records attributed to Donald Loveland.

At least 19 recordsLinked to original sources
arXiv2026

Training-Free LLM-Based Recommendation with Post-LLM Item Refinement Using Collaborative Signals

Large language models (LLMs) have shown promise for training-free recommendation, but LLM-generated user interests are often too broad for fine-grained item retrieval. Existing methods incorporate collaborative filtering (CF) signals in a pre-LLM manner through candidate reranking or prompt augmentation, yielding limited gains. We propose CoRRe, a training-free recommendation framework with a post-LLM paradigm that injects CF signals into LLM-generated item representations, which are later matched with LLM-generated user interests for ranking. Specifically, CoRRe refines the directions of item embeddings using an item-item co-purchase graph and their magnitudes using item popularity. Experiments on real-world datasets show that CoRRe consistently outperforms existing training-free methods and achieves competitive or superior performance compared with training-based methods, without requiring any model training or task-specific fine-tuning.

cs.IR↗
arXiv2026

LLM-Based Generative Retrieval for Snapchat Content Recommendation

Pretrained large language models (LLMs) are promising retrieval engines because they combine rich semantic priors, strong sequence modeling capabilities, and favorable scaling behavior. However, turning a pretrained LLM into a generative retriever in production deployment raises several challenges: the model must learn an internal item vocabulary that was absent from pretraining, and generate valid item identifiers under strict latency and cost constraints. We address these challenges through the design and launch of SnapLGR, an LLM-based generative retrieval system for short-video recommendation at Snapchat. The system is built around three main designs. First, we construct semantic identifiers (SIDs) from multimodal item embeddings and enhance them with Personalized PageRank (PPR)-based co-engagement contrastive learning, resulting in improved codebook utilization, reduced collisions, and infused collaborative signal. Second, we use continued pretraining (CPT) to ground the introduced SID tokens before supervised fine-tuning (SFT) on user interaction sequences. Third, we make SnapLGR serving practical through TensorRT-LLM CUDA-backed beam search and a decentralized worker-loop architecture. In a live A/B test, the launched system increased View Time by 0.37%, Time Spent by 0.09%, Deep Sessions by 0.18%, and Deep Sessions Unique User by 0.11% relative to the existing TIGER-style generative retrieval baseline. We then decompose this offline gap under a fixed tokenizer and quantify the gains due to model architecture, scaling, and pretraining. Overall, our deployment shows that successful production SnapLGR requires joint design across representation learning, vocabulary grounding, and efficient training and serving.

cs.IR↗
arXiv2026

Preserving Item Semantics for Free: Rethinking Token Initialization in LLM-Based Generative Recommendation

Recent advances in generative recommendation (GR) leverage large language models (LLMs) as recommender backbones, enabling LLMs to directly generate recommendations conditioned on item-interaction histories. In these systems, items are often represented through semantic IDs (SIDs) added to the LLM vocabulary as special tokens. Ideally, SIDs imbue item token representations with semantic priors, thereby improving model generalization. However, standard vocabulary expansion typically initializes these tokens as random Gaussian vectors, discarding the SIDs' underlying continuous geometry and forcing the LLM to relearn token relationships from interaction data. To demonstrate the consequences of this design, we first show that training from this initialization tends to organize SID embeddings around item popularity rather than semantics. We further show that, despite partially reducing the reliance on popularity and improving cold item performance, the computationally expensive process of continual pretraining (CPT) fails to reliably recover the original semantic geometry. To address these findings, we propose a simple, parameter-free intervention that initializes SID token embeddings directly from their corresponding centroids in the semantic embedding space. Requiring only a few lines of code and no additional training or inference overhead, this drop-in approach improves pure-SFT Recall@5 by up to 16%, reaches peak performance with up to 40% fewer SFT steps, and improves cold-item Recall@5 by up to 60%. Moreover, on datasets that benefit from additional CPT, centroid initialization reaches comparable performance while requiring half as many CPT epochs. Together, our findings show that preserving SID geometry, beyond shared-prefix structure, provides a simple and effective semantic prior for LLM-based GR.

cs.IR↗
arXiv2026

Implicit Reasoning for Large Language Model-based Generative Recommendation

Large Language Models (LLMs) are increasingly adopted as backbones for Generative Recommendation (GR), promising access to pretrained world knowledge. Yet reliably invoking this knowledge for GR remains poorly understood. A key obstacle is that LLM-based GR typically represents items with Semantic IDs (SIDs), disrupting LLMs' natural-language reasoning interface because these tokens are unseen by the LLM during pretraining. Existing approaches address this with expensive multi-stage pipelines that ground SIDs and elicit explicit rationales, but offer limited insight into when and why each stage is necessary. In this work, we systematically decompose explicit reasoning training pipelines for LLM-based GR, revealing three key limitations: weakened world-knowledge verbalization, misalignment between SID and natural-language token embedding spaces, and sensitivity to rationale quality, all of which hurt explicit reasoning performance. To circumvent these issues, we propose PauseRec, a lightweight implicit reasoning paradigm tailored for GR. PauseRec is exceptionally practical, avoiding costly reasoning trace acquisition and reasoning alignment training, leading to a multitude of benefits: (1) it outperforms standard explicit CoT methods by up to 6.22%, (2) it reduces training cost by up to 65% GPU hours, and (3) it speeds up inference by up to 71.3%. These results position PauseRec as a lightweight alternative to explicit rationale generation, enabling more effective and efficient LLM-based GR.

cs.CL↗
arXiv2026

Plain Transformers are Surprisingly Powerful Link Predictors

Link prediction is a core challenge in graph machine learning, demanding models that capture rich and complex topological dependencies. While Graph Neural Networks (GNNs) are the standard solution, state-of-the-art pipelines often rely on explicit structural heuristics or memory-intensive node embeddings -- approaches that struggle to generalize or scale to massive graphs. Emerging Graph Transformers (GTs) offer a potential alternative but often incur significant overhead due to complex structural encodings, hindering their applications to large-scale link prediction. We challenge these sophisticated paradigms with PENCIL, an encoder-only plain Transformer that replaces hand-crafted priors with attention over sampled local subgraphs, retaining the scalability and hardware efficiency of standard Transformers. Through experimental and theoretical analysis, we show that PENCIL extracts richer structural signals than GNNs, implicitly generalizing a broad class of heuristics and subgraph-based expressivity. Empirically, PENCIL outperforms heuristic-informed GNNs and is far more parameter-efficient than ID-embedding--based alternatives, while remaining competitive across diverse benchmarks -- even without node features. Our results challenge the prevailing reliance on complex engineering techniques, demonstrating that simple design choices are potentially sufficient to achieve the same capabilities. Our code is publicly available at https://github.com/quang-truong/pencil.

cs.LG↗
arXiv2026

On the Memorization Behavior of LLMs in Generative Recommendation: Observations, Implications, and Training Strategies

Generative recommendation (GR) has emerged as a promising direction for recommender systems. Recently, large language models (LLMs) have been increasingly adopted for GR, as their rich pretrained knowledge is expected to help them generalize beyond common user behavior patterns that traditional memorization-oriented baselines can capture. However, existing LLM-based GR works largely ignore LLMs' well-known tendency to memorize, which, if present in LLMs fine-tuned for GR, would restrict their utilization of pretrained knowledge. In this work, we investigate this concern by examining one-hop memorization, where a model recommends items that are direct successors of items in the training data. We show that LLMs do this more than non-LLM-based GR models-in fact, the vast majority of their gains over GR baselines are actually on users whose target items can be predicted through one-hop memorization. We intuit that improving performance on the remaining users requires LLMs to learn richer item-item relations beyond one-hop transitions. To achieve this, we propose IIRG, a novel training strategy that teaches LLMs to capture: (1) collaborative relations derived from item co-occurrences across multiple hops in user sequences, and (2) semantic relations among items with similar themes, both of which can serve as useful recommendation signals. We show that IIRG significantly improves over LLMs trained solely with standard next-item prediction, with especially large gains for users whose test items are not covered by train-time one-hop transitions.

cs.IR↗
arXiv2026

Formalizing and Mitigating Structural Distortion in LLM Attention for Graph Reasoning

Large Language Models (LLMs) have shown promise for reasoning over Text-Attributed Graphs (TAGs). However, applying LLMs to graphs requires linearizing their structure into sequences, introducing distortion rooted in the graph bandwidth problem. While this distortion has been shown to degrade performance, it is often attributed to prompt design or model scale, leaving the underlying mechanism unclear. In this work, we show \textit{how} rotary positional embeddings turn graph linearization into bandwidth-dependent attention decay, suppressing attention between graph-adjacent nodes that are forced far apart in the serialized sequence. This shifts the focus of LLM-based graph reasoning from prompt engineering and scaling toward correcting attention misalignment. Motivated by this analysis, we propose \textbf{G}raph-\textbf{a}ligned \textbf{L}anguage \textbf{A}ttention (\textbf{GaLA}), a lightweight, inference-time modification for LLMs. GaLA biases attention toward graph-adjacent nodes while preserving the LLM's sequential inductive biases. Across TAG benchmarks, GaLA improves performance with negligible overhead, demonstrating that distortion is a correctable bottleneck in LLM-based graph reasoning.

cs.LG↗
arXiv2026

Exploiting ID-Text Complementarity via Ensembling for Sequential Recommendation

Modern Sequential Recommendation (SR) models commonly utilize modality features to represent items, motivated in large part by recent advancements in language and vision modeling. To do so, several works completely replace ID embeddings with modality embeddings, claiming that modality embeddings render ID embeddings unnecessary because they can match or even exceed ID embedding performance. On the other hand, many works jointly utilize ID and modality features, but posit that complex fusion strategies, such as multi-stage training and/or intricate alignment architectures, are necessary for this joint utilization. However, underlying both these lines of work is a lack of understanding of the complementarity of ID and modality features. In this work, we address this gap by studying the complementarity of ID- and text-based SR models. We show that these models do learn complementary signals, meaning that either should provide performance gain when used properly alongside the other. Motivated by this, we propose a new SR method that preserves ID-text complementarity through independent model training, then harnesses it through a simple ensembling strategy. Despite this method's simplicity, we show it outperforms several competitive SR baselines, implying that both ID and text features are necessary to achieve state-of-the-art SR performance but complex fusion architectures are not.

cs.LG↗
arXiv2025

Glance for Context: Learning When to Leverage LLMs for Node-Aware GNN-LLM Fusion

Learning on text-attributed graphs has motivated the use of Large Language Models (LLMs) for graph learning. However, most fusion strategies are applied uniformly across all nodes and attain only small overall performance gains. We argue this result stems from aggregate metrics that obscure when LLMs provide benefit, inhibiting actionable signals for new strategies. In this work, we reframe LLM-GNN fusion around nodes where GNNs typically falter. We first show that performance can significantly differ between GNNs and LLMs, with each excelling on distinct structural patterns, such as local homophily. To leverage this finding, we propose GLANCE (GNN with LLM Assistance for Neighbor- and Context-aware Embeddings), a framework that invokes an LLM to refine a GNN's prediction. GLANCE employs a lightweight router that, given inexpensive per-node signals, decides whether to query the LLM. Since the LLM calls are non-differentiable, the router is trained with an advantage-based objective that compares the utility of querying the LLM against relying solely on the GNN. Across multiple benchmarks, GLANCE achieves the best performance balance across node subgroups, achieving significant gains on heterophilous nodes (up to $+13\%$) while simultaneously achieving top overall performance. Our findings highlight the value of adaptive, node-aware GNN-LLM architectures, where selectively invoking the LLM enables scalable deployment on large graphs without incurring high computational costs.

cs.LG↗
arXiv2025

Understanding and Scaling Collaborative Filtering Optimization from the Perspective of Matrix Rank

Collaborative Filtering (CF) methods dominate real-world recommender systems given their ability to learn high-quality, sparse ID-embedding tables that effectively capture user preferences. These tables scale linearly with the number of users and items, and are trained to ensure high similarity between embeddings of interacted user-item pairs, while maintaining low similarity for non-interacted pairs. Despite their high performance, encouraging dispersion for non-interacted pairs necessitates expensive regularization (e.g., negative sampling), hurting runtime and scalability. Existing research tends to address these challenges by simplifying the learning process, either by reducing model complexity or sampling data, trading performance for runtime. In this work, we move beyond model-level modifications and study the properties of the embedding tables under different learning strategies. Through theoretical analysis, we find that the singular values of the embedding tables are intrinsically linked to different CF loss functions. These findings are empirically validated on real-world datasets, demonstrating the practical benefits of higher stable rank, a continuous version of matrix rank which encodes the distribution of singular values. Based on these insights, we propose an efficient warm-start strategy that regularizes the stable rank of the user and item embeddings. We show that stable rank regularization during early training phases can promote higher-quality embeddings, resulting in training speed improvements of up to 66%. Additionally, stable rank regularization can act as a proxy for negative sampling, allowing for performance gains of up to 21% over loss functions with small negative sampling ratios. Overall, our analysis unifies current CF methods under a new perspective, their optimization of stable rank, motivating a flexible regularization method.

cs.IR↗
arXiv2025

On the Role of Weight Decay in Collaborative Filtering: A Popularity Perspective

Collaborative filtering (CF) enables large-scale recommendation systems by encoding information from historical user-item interactions into dense ID-embedding tables. However, as embedding tables grow, closed-form solutions become impractical, often necessitating the use of mini-batch gradient descent for training. Despite extensive work on designing loss functions to train CF models, we argue that one core component of these pipelines is heavily overlooked: weight decay. Attaining high-performing models typically requires careful tuning of weight decay, regardless of loss, yet its necessity is not well understood. In this work, we question why weight decay is crucial in CF pipelines and how it impacts training. Through theoretical and empirical analysis, we surprisingly uncover that weight decay's primary function is to encode popularity information into the magnitudes of the embedding vectors. Moreover, we find that tuning weight decay acts as a coarse, non-linear knob to influence preference towards popular or unpopular items. Based on these findings, we propose PRISM (Popularity-awaRe Initialization Strategy for embedding Magnitudes), a straightforward yet effective solution to simplify the training of high-performing CF models. PRISM pre-encodes the popularity information typically learned through weight decay, eliminating its necessity. Our experiments show that PRISM improves performance by up to 4.77% and reduces training times by 38.48%, compared to state-of-the-art training strategies. Additionally, we parameterize PRISM to modulate the initialization strength, offering a cost-effective and meaningful strategy to mitigate popularity bias.

cs.IR↗
arXiv2025

GraphHash: Graph Clustering Enables Parameter Efficiency in Recommender Systems

Deep recommender systems rely heavily on large embedding tables to handle high-cardinality categorical features such as user/item identifiers, and face significant memory constraints at scale. To tackle this challenge, hashing techniques are often employed to map multiple entities to the same embedding and thus reduce the size of the embedding tables. Concurrently, graph-based collaborative signals have emerged as powerful tools in recommender systems, yet their potential for optimizing embedding table reduction remains unexplored. This paper introduces GraphHash, the first graph-based approach that leverages modularity-based bipartite graph clustering on user-item interaction graphs to reduce embedding table sizes. We demonstrate that the modularity objective has a theoretical connection to message-passing, which provides a foundation for our method. By employing fast clustering algorithms, GraphHash serves as a computationally efficient proxy for message-passing during preprocessing and a plug-and-play graph-based alternative to traditional ID hashing. Extensive experiments show that GraphHash substantially outperforms diverse hashing baselines on both retrieval and click-through-rate prediction tasks. In particular, GraphHash achieves on average a 101.52% improvement in recall when reducing the embedding table size by more than 75%, highlighting the value of graph-based collaborative information for model reduction. Our code is available at https://github.com/snap-research/GraphHash.

cs.IR↗
arXiv2024

Unveiling the Impact of Local Homophily on GNN Fairness: In-Depth Analysis and New Benchmarks

Graph Neural Networks (GNNs) often struggle to generalize when graphs exhibit both homophily (same-class connections) and heterophily (different-class connections). Specifically, GNNs tend to underperform for nodes with local homophily levels that differ significantly from the global homophily level. This issue poses a risk in user-centric applications where underrepresented homophily levels are present. Concurrently, fairness within GNNs has received substantial attention due to the potential amplification of biases via message passing. However, the connection between local homophily and fairness in GNNs remains underexplored. In this work, we move beyond global homophily and explore how local homophily levels can lead to unfair predictions. We begin by formalizing the challenge of fair predictions for underrepresented homophily levels as an out-of-distribution (OOD) problem. We then conduct a theoretical analysis that demonstrates how local homophily levels can alter predictions for differing sensitive attributes. We additionally introduce three new GNN fairness benchmarks, as well as a novel semi-synthetic graph generator, to empirically study the OOD problem. Across extensive analysis we find that two factors can promote unfairness: (a) OOD distance, and (b) heterophilous nodes situated in homophilous graphs. In cases where these two conditions are met, fairness drops by up to 24% on real world datasets, and 30% in semi-synthetic datasets. Together, our theoretical insights, empirical analysis, and algorithmic contributions unveil a previously overlooked source of unfairness rooted in the graph's homophily information.

cs.LG↗
arXiv2023

On Performance Discrepancies Across Local Homophily Levels in Graph Neural Networks

Graph Neural Network (GNN) research has highlighted a relationship between high homophily (i.e., the tendency of nodes of the same class to connect) and strong predictive performance in node classification. However, recent work has found the relationship to be more nuanced, demonstrating that simple GNNs can learn in certain heterophilous settings. To resolve these conflicting findings and align closer to real-world datasets, we go beyond the assumption of a global graph homophily level and study the performance of GNNs when the local homophily level of a node deviates from the global homophily level. Through theoretical and empirical analysis, we systematically demonstrate how shifts in local homophily can introduce performance degradation, leading to performance discrepancies across local homophily levels. We ground the practical implications of this work through granular analysis on five real-world datasets with varying global homophily levels, demonstrating that (a) GNNs can fail to generalize to test nodes that deviate from the global homophily of a graph, and (b) high local homophily does not necessarily confer high performance for a node. We further show that GNNs designed for globally heterophilous graphs can alleviate performance discrepancy by improving performance across local homophily levels, offering a new perspective on how these GNNs achieve stronger global performance.

cs.SI↗
arXiv2023

Network Design through Graph Neural Networks: Identifying Challenges and Improving Performance

Graph Neural Network (GNN) research has produced strategies to modify a graph's edges using gradients from a trained GNN, with the goal of network design. However, the factors which govern gradient-based editing are understudied, obscuring why edges are chosen and if edits are grounded in an edge's importance. Thus, we begin by analyzing the gradient computation in previous works, elucidating the factors that influence edits and highlighting the potential over-reliance on structural properties. Specifically, we find that edges can achieve high gradients due to structural biases, rather than importance, leading to erroneous edits when the factors are unrelated to the design task. To improve editing, we propose ORE, an iterative editing method that (a) edits the highest scoring edges and (b) re-embeds the edited graph to refresh gradients, leading to less biased edge choices. We empirically study ORE through a set of proposed design tasks, each with an external validation method, demonstrating that ORE improves upon previous methods by up to 50%.

cs.LG↗
arXiv2022

On Graph Neural Network Fairness in the Presence of Heterophilous Neighborhoods

We study the task of node classification for graph neural networks (GNNs) and establish a connection between group fairness, as measured by statistical parity and equal opportunity, and local assortativity, i.e., the tendency of linked nodes to have similar attributes. Such assortativity is often induced by homophily, the tendency for nodes of similar properties to connect. Homophily can be common in social networks where systemic factors have forced individuals into communities which share a sensitive attribute. Through synthetic graphs, we study the interplay between locally occurring homophily and fair predictions, finding that not all node neighborhoods are equal in this respect -- neighborhoods dominated by one category of a sensitive attribute often struggle to obtain fair treatment, especially in the case of diverging local class and sensitive attribute homophily. After determining that a relationship between local homophily and fairness exists, we investigate if the issue of unfairness can be associated to the design of the applied GNN model. We show that by adopting heterophilous GNN designs capable of handling disassortative group labels, group fairness in locally heterophilous neighborhoods can be improved by up to 25% over homophilous designs in real and synthetic datasets.

cs.SI↗
arXiv2022

How does Heterophily Impact the Robustness of Graph Neural Networks? Theoretical Connections and Practical Implications

We bridge two research directions on graph neural networks (GNNs), by formalizing the relation between heterophily of node labels (i.e., connected nodes tend to have dissimilar labels) and the robustness of GNNs to adversarial attacks. Our theoretical and empirical analyses show that for homophilous graph data, impactful structural attacks always lead to reduced homophily, while for heterophilous graph data the change in the homophily level depends on the node degrees. These insights have practical implications for defending against attacks on real-world graphs: we deduce that separate aggregators for ego- and neighbor-embeddings, a design principle which has been identified to significantly improve prediction for heterophilous graph data, can also offer increased robustness to GNNs. Our comprehensive experiments show that GNNs merely adopting this design achieve improved empirical and certifiable robustness compared to the best-performing unvaccinated model. Additionally, combining this design with explicit defense mechanisms against adversarial attacks leads to an improved robustness with up to 18.33% performance increase under attacks compared to the best-performing vaccinated model.

cs.LG↗
arXiv2022

Zeroth-Order SciML: Non-intrusive Integration of Scientific Software with Deep Learning

Using deep learning (DL) to accelerate and/or improve scientific workflows can yield discoveries that are otherwise impossible. Unfortunately, DL models have yielded limited success in complex scientific domains due to large data requirements. In this work, we propose to overcome this issue by integrating the abundance of scientific knowledge sources (SKS) with the DL training process. Existing knowledge integration approaches are limited to using differentiable knowledge source to be compatible with first-order DL training paradigm. In contrast, our proposed approach treats knowledge source as a black-box in turn allowing to integrate virtually any knowledge source. To enable an end-to-end training of SKS-coupled-DL, we propose to use zeroth-order optimization (ZOO) based gradient-free training schemes, which is non-intrusive, i.e., does not require making any changes to the SKS. We evaluate the performance of our ZOO training scheme on two real-world material science applications. We show that proposed scheme is able to effectively integrate scientific knowledge with DL training and is able to outperform purely data-driven model for data-limited scientific applications. We also discuss some limitations of the proposed method and mention potentially worthwhile future directions.

cs.LG↗