arXiv subjects

Maarten de Rijke

Publications and source records attributed to Maarten de Rijke.

At least 73 records · Page 4Linked to original sources

arXiv2025

A Reproducibility Study of Product-side Fairness in Bundle Recommendation

Recommender systems are known to exhibit fairness issues, particularly on the product side, where products and their associated suppliers receive unequal exposure in recommended results. While this problem has been widely studied in traditional recommendation settings, its implications for bundle recommendation (BR) remain largely unexplored. This emerging task introduces additional complexity: recommendations are generated at the bundle level, yet user satisfaction and product (or supplier) exposure depend on both the bundle and the individual items it contains. Existing fairness frameworks and metrics designed for traditional recommender systems may not directly translate to this multi-layered setting. In this paper, we conduct a comprehensive reproducibility study of product-side fairness in BR across three real-world datasets using four state-of-the-art BR methods. We analyze exposure disparities at both the bundle and item levels using multiple fairness metrics, uncovering important patterns. Our results show that exposure patterns differ notably between bundles and items, revealing the need for fairness interventions that go beyond bundle-level assumptions. We also find that fairness assessments vary considerably depending on the metric used, reinforcing the need for multi-faceted evaluation. Furthermore, user behavior plays a critical role: when users interact more frequently with bundles than with individual items, BR systems tend to yield fairer exposure distributions across both levels. Overall, our findings offer actionable insights for building fairer bundle recommender systems and establish a vital foundation for future research in this emerging domain.

cs.IR↗

arXiv2025

Evaluation of Attribution Bias in Generator-Aware Retrieval-Augmented Large Language Models

Attributing answers to source documents is an approach used to enhance the verifiability of a model's output in retrieval augmented generation (RAG). Prior work has mainly focused on improving and evaluating the attribution quality of large language models (LLMs) in RAG, but this may come at the expense of inducing biases in the attribution of answers. We define and examine two aspects in the evaluation of LLMs in RAG pipelines, namely attribution sensitivity and bias with respect to authorship information. We explicitly inform an LLM about the authors of source documents, instruct it to attribute its answers, and analyze (i) how sensitive the LLM's output is to the author of source documents, and (ii) whether the LLM exhibits a bias towards human-written or AI-generated source documents. We design an experimental setup in which we use counterfactual evaluation to study three LLMs in terms of their attribution sensitivity and bias in RAG pipelines. Our results show that adding authorship information to source documents can significantly change the attribution quality of LLMs by 3% to 18%. Moreover, we show that LLMs can have an attribution bias towards explicit human authorship, which can serve as a competing hypothesis for findings of prior work that shows that LLM-generated content may be preferred over human-written contents. Our findings indicate that metadata of source documents can influence LLMs' trust, and how they attribute their answers. Furthermore, our research highlights attribution bias and sensitivity as a novel aspect of brittleness in LLMs.

cs.CL↗

arXiv2025

Rethinking Click Models in Light of Carousel Interfaces: Theory-Based Categorization and Design of Click Models

Click models are a well-established for modeling user interactions with web interfaces. Previous work has mainly focused on traditional single-list web search settings; this includes existing surveys that introduced categorizations based on the first generation of probabilistic graphical model (PGM) click models that have become standard. However, these categorizations have become outdated, as their conceptualizations are unable to meaningfully compare PGM with neural network (NN) click models nor generalize to newer interfaces, such as carousel interfaces. We argue that this outdated view fails to adequately explain the fundamentals of click model designs, thus hindering the development of novel click models. This work reconsiders what should be the fundamental concepts in click model design, grounding them - unlike previous approaches - in their mathematical properties. We propose three fundamental key-design choices that explain what statistical patterns a click model can capture, and thus indirectly, what user behaviors they can capture. Based on these choices, we create a novel click model taxonomy that allows a meaningful comparison of all existing click models; this is the first taxonomy of single-list, grid and carousel click models that includes PGMs and NNs. Finally, we show how our conceptualization provides a foundation for future click model design by an example derivation of a novel design for carousel interfaces.

cs.IR↗

arXiv2025

Unidentified and Confounded? Understanding Two-Tower Models for Unbiased Learning to Rank

Additive two-tower models are popular learning-to-rank methods for handling biased user feedback in industry settings. Recent studies, however, report a concerning phenomenon: training two-tower models on clicks collected by well-performing production systems leads to decreased ranking performance. This paper investigates two recent explanations for this observation: confounding effects from logging policies and model identifiability issues. We theoretically analyze the identifiability conditions of two-tower models, showing that either document swaps across positions or overlapping feature distributions are required to recover model parameters from clicks. We also investigate the effect of logging policies on two-tower models, finding that they introduce no bias when models perfectly capture user behavior. However, logging policies can amplify biases when models imperfectly capture user behavior, particularly when prediction errors correlate with document placement across positions. We propose a sample weighting technique to mitigate these effects and provide actionable insights for researchers and practitioners using two-tower models.

cs.IR↗

arXiv2025

Optimized Text Embedding Models and Benchmarks for Amharic Passage Retrieval

Neural retrieval methods using transformer-based pre-trained language models have advanced multilingual and cross-lingual retrieval. However, their effectiveness for low-resource, morphologically rich languages such as Amharic remains underexplored due to data scarcity and suboptimal tokenization. We address this gap by introducing Amharic-specific dense retrieval models based on pre-trained Amharic BERT and RoBERTa backbones. Our proposed RoBERTa-Base-Amharic-Embed model (110M parameters) achieves a 17.6% relative improvement in MRR@10 and a 9.86% gain in Recall@10 over the strongest multilingual baseline, Arctic Embed 2.0 (568M parameters). More compact variants, such as RoBERTa-Medium-Amharic-Embed (42M), remain competitive while being over 13x smaller. Additionally, we train a ColBERT-based late interaction retrieval model that achieves the highest MRR@10 score (0.843) among all evaluated models. We benchmark our proposed models against both sparse and dense retrieval baselines to systematically assess retrieval effectiveness in Amharic. Our analysis highlights key challenges in low-resource settings and underscores the importance of language-specific adaptation. To foster future research in low-resource IR, we publicly release our dataset, codebase, and trained models at https://github.com/kidist-amde/amharic-ir-benchmarks.

cs.IR↗

arXiv2025

Benchmark Granularity and Model Robustness for Image-Text Retrieval

Image-Text Retrieval (ITR) systems are central to multimodal information access, with Vision-Language Models (VLMs) showing strong performance on standard benchmarks. However, these benchmarks predominantly rely on coarse-grained annotations, limiting their ability to reveal how models perform under real-world conditions, where query granularity varies. Motivated by this gap, we examine how dataset granularity and query perturbations affect retrieval performance and robustness across four architecturally diverse VLMs (ALIGN, AltCLIP, CLIP, and GroupViT). Using both standard benchmarks (MS-COCO, Flickr30k) and their fine-grained variants, we show that richer captions consistently enhance retrieval, especially in text-to-image tasks, where we observe an average improvement of 16.23%, compared to 6.44% in image-to-text. To assess robustness, we introduce a taxonomy of perturbations and conduct extensive experiments, revealing that while perturbations typically degrade performance, they can also unexpectedly improve retrieval, exposing nuanced model behaviors. Notably, word order emerges as a critical factor -- contradicting prior assumptions of model insensitivity to it. Our results highlight variation in model robustness and a dataset-dependent relationship between caption granularity and perturbation sensitivity and emphasize the necessity of evaluating models on datasets of varying granularity.

cs.CV↗

arXiv2025

DeepShop: A Benchmark for Deep Research Shopping Agents

Web agents for online shopping have shown great promise in automating user interactions across e-commerce platforms. Benchmarks for assessing such agents do not reflect the complexity of real-world shopping scenarios, as they often consist of overly simple queries with deterministic paths, such as "Find iPhone 15." Real shopping scenarios are inherently more layered, involving multi-dimensional product attributes, search filters, and user-specific sorting preferences. To address this gap, we introduce DeepShop, a benchmark designed to evaluate web agents in complex and realistic online shopping environments. DeepShop comprises three key components. (1) Query diversity evolution: Starting from real user queries, we generate diverse queries across five popular online shopping domains. (2) Query complexity evolution: We further evolve these queries to increase complexity, considering product attributes, search filters, and sorting preferences, and classify them into three levels: easy, medium, and hard, based on the number of evolutions. (3) Fine-grained and holistic evaluation: We propose an automated evaluation framework that assesses agent performance in terms of fine-grained aspects (product attributes, search filters, and sorting preferences) and reports the overall success rate through holistic evaluation. We conduct a systematic evaluation of retrieval-augmented generation (RAG) methods, web agents, and deep research systems. Results show that RAG struggles with complex queries due to its lack of web interaction, while other methods face significant challenges with filters and sorting preferences, leading to low overall success rates. We also perform cross-category, complexity-based evaluations and error analyses to support the advancement of deep research shopping agents.

cs.IR↗

arXiv2025

On the Scaling of Robustness and Effectiveness in Dense Retrieval

Robustness and Effectiveness are critical aspects of developing dense retrieval models for real-world applications. It is known that there is a trade-off between the two. Recent work has addressed scaling laws of effectiveness in dense retrieval, revealing a power-law relationship between effectiveness and the size of models and data. Does robustness follow scaling laws too? If so, can scaling improve both robustness and effectiveness together, or do they remain locked in a trade-off? To answer these questions, we conduct a comprehensive experimental study. We find that:(i) Robustness, including out-of-distribution and adversarial robustness, also follows a scaling law.(ii) Robustness and effectiveness exhibit different scaling patterns, leading to significant resource costs when jointly improving both. Given these findings, we shift to the third factor that affects model performance, namely the optimization strategy, beyond the model size and data size. We find that: (i) By fitting different optimization strategies, the joint performance of robustness and effectiveness traces out a Pareto frontier. (ii) When the optimization strategy strays from Pareto efficiency, the joint performance scales in a sub-optimal direction. (iii) By adjusting the optimization weights to fit the Pareto efficiency, we can achieve Pareto training, where the scaling of joint performance becomes most efficient. Even without requiring additional resources, Pareto training is comparable to the performance of scaling resources several times under optimization strategies that overly prioritize either robustness or effectiveness. Finally, we demonstrate that our findings can help deploy dense retrieval models in real-world applications that scale efficiently and are balanced for robustness and effectiveness.

cs.IR↗

arXiv2025

The Silent Saboteur: Imperceptible Adversarial Attacks against Black-Box Retrieval-Augmented Generation Systems

We explore adversarial attacks against retrieval-augmented generation (RAG) systems to identify their vulnerabilities. We focus on generating human-imperceptible adversarial examples and introduce a novel imperceptible retrieve-to-generate attack against RAG. This task aims to find imperceptible perturbations that retrieve a target document, originally excluded from the initial top-$k$ candidate set, in order to influence the final answer generation. To address this task, we propose ReGENT, a reinforcement learning-based framework that tracks interactions between the attacker and the target RAG and continuously refines attack strategies based on relevance-generation-naturalness rewards. Experiments on newly constructed factual and non-factual question-answering benchmarks demonstrate that ReGENT significantly outperforms existing attack methods in misleading RAG systems with small imperceptible text perturbations.

cs.IR↗

arXiv2025

Distributional Reinforcement Learning with Dual Expectile-Quantile Regression

Distributional reinforcement learning (RL) has proven useful in multiple benchmarks as it enables approximating the full distribution of returns and extracts rich feedback from environment samples. The commonly used quantile regression approach to distributional RL -- based on asymmetric $L_1$ losses -- provides a flexible and effective way of learning arbitrary return distributions. In practice, it is often improved by using a more efficient, asymmetric hybrid $L_1$-$L_2$ Huber loss for quantile regression. However, by doing so, distributional estimation guarantees vanish, and we empirically observe that the estimated distribution rapidly collapses to its mean. Indeed, asymmetric $L_2$ losses, corresponding to expectile regression, cannot be readily used for distributional temporal difference. Motivated by the efficiency of $L_2$-based learning, we propose to jointly learn expectiles and quantiles of the return distribution in a way that allows efficient learning while keeping an estimate of the full distribution of returns. We prove that our proposed operator converges to the distributional Bellman operator in the limit of infinite estimated quantile and expectile fractions, and we benchmark a practical implementation on a toy example and at scale. On the Atari benchmark, our approach matches the performance of the Huber-based IQN-1 baseline after $200$M training frames but avoids distributional collapse and keeps estimates of the full distribution of returns.

cs.LG↗

arXiv2025

MELoRA: Mini-Ensemble Low-Rank Adapters for Parameter-Efficient Fine-Tuning

Parameter-efficient fine-tuning (PEFT) is a popular method for tailoring pre-trained large language models (LLMs), especially as the models' scale and the diversity of tasks increase. Low-rank adaptation (LoRA) is based on the idea that the adaptation process is intrinsically low-dimensional, i.e., significant model changes can be represented with relatively few parameters. However, decreasing the rank encounters challenges with generalization errors for specific tasks when compared to full-parameter fine-tuning. We present MELoRA, a mini-ensemble low-rank adapters that uses fewer trainable parameters while maintaining a higher rank, thereby offering improved performance potential. The core idea is to freeze original pretrained weights and train a group of mini LoRAs with only a small number of parameters. This can capture a significant degree of diversity among mini LoRAs, thus promoting better generalization ability. We conduct a theoretical analysis and empirical studies on various NLP tasks. Our experimental results show that, compared to LoRA, MELoRA achieves better performance with 8 times fewer trainable parameters on natural language understanding tasks and 36 times fewer trainable parameters on instruction following tasks, which demonstrates the effectiveness of MELoRA.

cs.CL↗

arXiv2025

Iterative Self-Incentivization Empowers Large Language Models as Agentic Searchers

Large language models (LLMs) have been widely integrated into information retrieval to advance traditional techniques. However, effectively enabling LLMs to seek accurate knowledge in complex tasks remains a challenge due to the complexity of multi-hop queries as well as the irrelevant retrieved content. To address these limitations, we propose EXSEARCH, an agentic search framework, where the LLM learns to retrieve useful information as the reasoning unfolds through a self-incentivized process. At each step, the LLM decides what to retrieve (thinking), triggers an external retriever (search), and extracts fine-grained evidence (recording) to support next-step reasoning. To enable LLM with this capability, EXSEARCH adopts a Generalized Expectation-Maximization algorithm. In the E-step, the LLM generates multiple search trajectories and assigns an importance weight to each; the M-step trains the LLM on them with a re-weighted loss function. This creates a self-incentivized loop, where the LLM iteratively learns from its own generated data, progressively improving itself for search. We further theoretically analyze this training process, establishing convergence guarantees. Extensive experiments on four knowledge-intensive benchmarks show that EXSEARCH substantially outperforms baselines, e.g., +7.8% improvement on exact match score. Motivated by these promising results, we introduce EXSEARCH-Zoo, an extension that extends our method to broader scenarios, to facilitate future work.

cs.CL↗

arXiv2025

Query Performance Prediction using Relevance Judgments Generated by Large Language Models

Query performance prediction (QPP) aims to estimate the retrieval quality of a search system for a query without human relevance judgments. Previous QPP methods typically return a single scalar value and do not require the predicted values to approximate a specific information retrieval (IR) evaluation measure, leading to certain drawbacks: (i) a single scalar is insufficient to accurately represent different IR evaluation measures, especially when metrics do not highly correlate, and (ii) a single scalar limits the interpretability of QPP methods because solely using a scalar is insufficient to explain QPP results. To address these issues, we propose a QPP framework using automatically generated relevance judgments (QPP-GenRE), which decomposes QPP into independent subtasks of predicting the relevance of each item in a ranked list to a given query. This allows us to predict any IR evaluation measure using the generated relevance judgments as pseudo-labels. This also allows us to interpret predicted IR evaluation measures, and identify, track and rectify errors in generated relevance judgments to improve QPP quality. We predict an item's relevance by using open-source large language models (LLMs) to ensure scientific reproducibility. We face two main challenges: (i) excessive computational costs of judging an entire corpus for predicting a metric considering recall, and (ii) limited performance in prompting open-source LLMs in a zero-/few-shot manner. To solve the challenges, we devise an approximation strategy to predict an IR measure considering recall and propose to fine-tune open-source LLMs using human-labeled relevance judgments. Experiments on the TREC 2019 to 2022 deep learning tracks and CAsT-19 and 20 datasets show that QPP-GenRE achieves state-of-the-art QPP quality for both lexical and neural rankers.

cs.IR↗

arXiv2025

Pretraining Data Detection for Large Language Models: A Divergence-based Calibration Method

As the scale of training corpora for large language models (LLMs) grows, model developers become increasingly reluctant to disclose details on their data. This lack of transparency poses challenges to scientific evaluation and ethical deployment. Recently, pretraining data detection approaches, which infer whether a given text was part of an LLM's training data through black-box access, have been explored. The Min-K\% Prob method, which has achieved state-of-the-art results, assumes that a non-training example tends to contain a few outlier words with low token probabilities. However, the effectiveness may be limited as it tends to misclassify non-training texts that contain many common words with high probabilities predicted by LLMs. To address this issue, we introduce a divergence-based calibration method, inspired by the divergence-from-randomness concept, to calibrate token probabilities for pretraining data detection. We compute the cross-entropy (i.e., the divergence) between the token probability distribution and the token frequency distribution to derive a detection score. We have developed a Chinese-language benchmark, PatentMIA, to assess the performance of detection approaches for LLMs on Chinese text. Experimental results on English-language benchmarks and PatentMIA demonstrate that our proposed method significantly outperforms existing methods. Our code and PatentMIA benchmark are available at https://github.com/zhang-wei-chao/DC-PDD.

cs.CL↗

arXiv2025

Direct Retrieval-augmented Optimization: Synergizing Knowledge Selection and Language Models

Retrieval-augmented generation (RAG) integrates large language models ( LLM s) with retrievers to access external knowledge, improving the factuality of LLM generation in knowledge-grounded tasks. To optimize the RAG performance, most previous work independently fine-tunes the retriever to adapt to frozen LLM s or trains the LLMs to use documents retrieved by off-the-shelf retrievers, lacking end-to-end training supervision. Recent work addresses this limitation by jointly training these two components but relies on overly simplifying assumptions of document independence, which has been criticized for being far from real-world scenarios. Thus, effectively optimizing the overall RAG performance remains a critical challenge. We propose a direct retrieval-augmented optimization framework, named DRO, that enables end-to-end training of two key components: (i) a generative knowledge selection model and (ii) an LLM generator. DRO alternates between two phases: (i) document permutation estimation and (ii) re-weighted maximization, progressively improving RAG components through a variational approach. In the estimation step, we treat document permutation as a latent variable and directly estimate its distribution from the selection model by applying an importance sampling strategy. In the maximization step, we calibrate the optimization expectation using importance weights and jointly train the selection model and LLM generator. Our theoretical analysis reveals that DRO is analogous to policy-gradient methods in reinforcement learning. Extensive experiments conducted on five datasets illustrate that DRO outperforms the best baseline with 5%-15% improvements in EM and F1. We also provide in-depth experiments to qualitatively analyze the stability, convergence, and variance of DRO.

cs.IR↗

arXiv2025

RankingSHAP -- Listwise Feature Attribution Explanations for Ranking Models

While SHAP (SHapley Additive exPlanations) and other feature attribution methods are commonly employed to explain model predictions, their application within information retrieval (IR), particularly for complex outputs such as ranked lists, remains limited. Existing attribution methods typically provide pointwise explanations, focusing on why a single document received a high-ranking score, rather than considering the relationships between documents in a ranked list. We present three key contributions to address this gap. First, we rigorously define listwise feature attribution for ranking models. Secondly, we introduce RankingSHAP, extending the popular SHAP framework to accommodate listwise ranking attribution, addressing a significant methodological gap in the field. Third, we propose two novel evaluation paradigms for assessing the faithfulness of attributions in learning-to-rank models, measuring the correctness and completeness of the explanation with respect to different aspects. Through experiments on standard learning-to-rank datasets, we demonstrate RankingSHAP's practical application while identifying the constraints of selection-based explanations. We further employ a simulated study with an interpretable model to showcase how listwise ranking attributions can be used to examine model decisions and conduct a qualitative evaluation of explanations. Due to the contrastive nature of the ranking task, our understanding of ranking model decisions can substantially benefit from feature attribution explanations like RankingSHAP.

cs.IR↗

arXiv2025

RecGaze: The First Eye Tracking and User Interaction Dataset for Carousel Interfaces

Carousel interfaces are widely used in e-commerce and streaming services, but little research has been devoted to them. Previous studies of interfaces for presenting search and recommendation results have focused on single ranked lists, but it appears their results cannot be extrapolated to carousels due to the added complexity. Eye tracking is a highly informative approach to understanding how users click, yet there are no eye tracking studies concerning carousels. There are very few interaction datasets on recommenders with carousel interfaces and none that contain gaze data. We introduce the RecGaze dataset: the first comprehensive feedback dataset on carousels that includes eye tracking results, clicks, cursor movements, and selection explanations. The dataset comprises of interactions from 3 movie selection tasks with 40 different carousel interfaces per user. In total, 87 users and 3,477 interactions are logged. In addition to the dataset, its description and possible use cases, we provide results of a survey on carousel design and the first analysis of gaze data on carousels, which reveals a golden triangle or F-pattern browsing behavior. Our work seeks to advance the field of carousel interfaces by providing the first dataset with eye tracking results on carousels. In this manner, we provide and encourage an empirical understanding of interactions with carousel interfaces, for building better recommender systems through gaze information, and also encourage the development of gaze-based recommenders.

cs.IR↗

arXiv2025

An Empirical Study of Evaluating Long-form Question Answering

\Ac{LFQA} aims to generate lengthy answers to complex questions. This scenario presents great flexibility as well as significant challenges for evaluation. Most evaluations rely on deterministic metrics that depend on string or n-gram matching, while the reliability of large language model-based evaluations for long-form answers remains relatively unexplored. We address this gap by conducting an in-depth study of long-form answer evaluation with the following research questions: (i) To what extent do existing automatic evaluation metrics serve as a substitute for human evaluations? (ii) What are the limitations of existing evaluation metrics compared to human evaluations? (iii) How can the effectiveness and robustness of existing evaluation methods be improved? We collect 5,236 factoid and non-factoid long-form answers generated by different large language models and conduct a human evaluation on 2,079 of them, focusing on correctness and informativeness. Subsequently, we investigated the performance of automatic evaluation metrics by evaluating these answers, analyzing the consistency between these metrics and human evaluations. We find that the style, length of the answers, and the category of questions can bias the automatic evaluation metrics. However, fine-grained evaluation helps mitigate this issue on some metrics. Our findings have important implications for the use of large language models for evaluating long-form question answering. All code and datasets are available at https://github.com/bugtig6351/lfqa_evaluation.

cs.IR↗