Peer Revenue Impact factor: volatile

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Computer Science Information Retrieval

cs.IR Watch Information Retrieval

Wednesday 12 August

Information Retrieval

From Overlooked to Explored: Recovering Item Relations via Mixture of Perspectives for Sequential Recommendation

Junyoung Kim, Wonbin Kweon, Woojoo Kim et al. (+3 more)

Capturing user preference from a user's interaction sequence is the central challenge of Sequential Recommendation (SR). This preference intuitively emerges from inter-item relations: each item transition reflects a preference embedded in the relations between items, making the faithful capture of these relations essential for accurate recommendation. For this reason, self-attention is dominant in sequential recommendation for its ability to compute pairwise item interactions, yet our empirical analysis reveals that it consistently suffers from similarity bias across various types of transformer-based SR models: dot-product attention scores disproportionately favor similar items, systematically overlooking heterogeneous relations with meaningful preference signals and directly limiting recommendation performance. To address this, we propose PRISM (Perspective-based Relational Insight Synthesis Module), a module that re-examines item relations from multiple perspectives. PRISM employs K Perspective Lenses to calibrate attention from distinct viewpoints, combining an Affinity View that refines homogeneous relations and a Contrast View that exposes heterogeneous ones suppressed by similarity bias, enabling the model to capture the full spectrum of user preferences. Extensive experiments on seven real-world benchmarks demonstrate that PRISM consistently outperforms state-of-the-art baselines. Our code is available at https://github.com/327aem/PRISM/.
Read the full abstract

All authors Junyoung Kim, Wonbin Kweon, Woojoo Kim, Jaehyung Lim, Dongha Kim, Hwanjo Yu

Information Retrieval · Cross-listed from cs.AI

AgenticTwin: An Agentic LLM Framework Integrated with Digital Twin for Anomaly Detection

Touseef Hasan, Mounika Ghanta, Souvika Sarkar et al. (+1 more)

Digital twins are increasingly used to monitor and simulate the behavior of cyber-physical systems. Even with skilled operators, interpreting anomalies detected within digital twin pipelines is challenging, as the sheer complexity and volume of raw sensor data make thorough analysis difficult. Recent advances in large language models (LLMs) offer promising capabilities for reasoning and explanation, yet their integration into digital twin-driven anomaly analysis remains underexplored. In this work, we propose AgenticTwin, an agentic framework that integrates LLM-driven reasoning with a digital twin-based anomaly detection pipeline. The framework grounds LLM-generated explanations in outputs from a digital twin-driven anomaly classifier and enables human operators to ask relevant natural-language questions about the system. Beyond the framework itself, we introduce a benchmark-oriented evaluation pipeline constructed over synthetic anomalies injected into a real-world weather sensor dataset, enabling controlled generation of operator queries over anomaly events. We further evaluate the feasibility of deploying lightweight, open-source LLMs for practical cyber-physical environments. Experimental results demonstrate that structured agent collaboration and knowledge-grounded reasoning improve diagnosis quality, contextual retrieval, and mitigation quality across diverse possible anomaly scenarios.
Read the full abstract

All authors Touseef Hasan, Mounika Ghanta, Souvika Sarkar, Ujjwal Guin

Information Retrieval · Cross-listed from cs.LG

FunnelCausalNet: Funnel-aware Joint Conversion-Revenue Uplift for Multi-tier Coupon Allocation

Yu Zhang, Zhihan Wang, Guanlin Chen et al. (+2 more)

Coupon campaigns seek to lift both conversion and revenue, but gross merchandise value (GMV) follows a deterministic funnel from conversion to conditional order value and is zero-inflated and heavy-tailed. We propose FunnelCausalNet, an uplift estimator coupling a binary conversion head with a nonnegative conditional-value head through $\mu_{\mathrm{gmv}}=\mu_{\mathrm{conv}}\mu_{\mathrm{val}}$. Under explicit RCT, support, rate-gap, and cross-head covariance-control assumptions, an idealized leading-order MSE comparison identifies a regime in which funnel composition can reduce pointwise variance; this is a heuristic, not a guarantee for the shared-representation neural model. The estimator is paired with marginal split-conformal CATE summaries, combined through a Bonferroni union as audit bands, and a Lagrangian budgeted allocator using RCT-anchored estimates for subsidy-aware ROI accounting. On semi-synthetic multi-tier Criteo-MT7, FunnelCausalNet's mean AUUC_GMV is within one seed standard deviation of the leading feature-interaction baseline among eleven baselines, while a controlled ablation reduces GMV effect error versus direct GMV regression by 18–48% across tested zero-inflation regimes. On de-identified industrial Hotel-Coupon RCT logs with about 4.9 million hold-out exposure records per seed, expected-outcome evaluation sweeps full LP frontiers; FunnelCausalNet has the best seed-averaged mean DeltaROI at all seven correlated anchors from 10% to 60%, which we treat as descriptive frontier consistency rather than independent significance. On sparse binary-spend public benchmarks, revenue-focused rankers can dominate uplift-curve proxies, defining an explicit regime boundary.
Read the full abstract

All authors Yu Zhang, Zhihan Wang, Guanlin Chen, Min Jiang, Shuai Li

Tuesday 11 August

Information Retrieval

Soft Curriculum Learning for Optimizing Fresh and Generalized Recommendations

Arnab Bhadury, Siyan Zheng, Anlan Yu et al. (+6 more)

Large-scale recommender systems, particularly short-form video platforms, are often bottlenecked by massive popularity feedback loops. In such environments, as models recommend popular items, they generate an overwhelming amount of skewed training data for "head" items. This creates a self-reinforcing cycle where retrieval and ranking models memorize "head" item patterns at the expense of generalizing across the vast "tail" of the catalogue. While Curriculum Learning (CL) offers a powerful mechanism to break this feedback loop by systematically exposing models to progressively more difficult and less frequent examples, its adoption in industrial recommendation has been hampered by hardware utilization inefficiencies or the needs for complicated pre-processing techniques because dynamic data rejection algorithms tend to starve hardware accelearators (TPUs/GPUs) by becoming largely CPU-bound. In this work, we introduce a scalable Soft Curriculum Learning framework designed specifically for continuous training setups within industry-scale retrieval and ranking models. By utilizing loss annealing and in-graph weight adjustments rather than rigid data filtering, we break the popularity feedback and enable dynamic curriculum pacing without sacrificing system throughput. We demonstrate empirical evidence through applications across sequence-based retrieval models (such as SASRec), two-tower retrieval models, and large-scale continuous ranking models. Online A/B tests on our short-video platform demonstrate substantial lifts in both overall user satisfaction and fresh content consumption, all without degrading model throughput.
Read the full abstract

All authors Arnab Bhadury, Siyan Zheng, Anlan Yu, Palaksh Rungta, Jiawei Li, Changping Meng, Dapeng Hong, Chuan He, Onkar Dalal

Information Retrieval

A Semantic Approach to the Academic Publishing Network: Document Vector Representations and Hybrid Structural-Semantic Fusion over OpenAlex Data

Robert Šamárek and Radek Martinek

Structural graph analysis of the academic publishing network captures the topological relationships between entities but does not see the content of works. Building on our structural approach, this work complements it with a semantic layer and a parameterized structural-semantic fusion. We represent scientific documents by citation-informed vector embeddings (SPECTER2) and store them in an embedded vector database keyed by the stable OpenAlex ID, so that they connect directly to the graph layer. We define a modular late-fusion function that combines semantic similarity (cosine of embeddings) and structural similarity (bibliographic coupling) with a tunable weight alpha whose value is chosen according to the specific task. On the corpus of VSB - Technical University of Ostrava we show two things: citation-informed embeddings agree with the expert OpenAlex topical taxonomy better than a TF-IDF baseline, and in a recommendation use case the structural, semantic, and combined signals carry information in different regimes depending on the available data. Hybrid fusion here is not a universally better method but an explicit mechanism for steering complementary signals according to the task. We release the whole approach as an open-source extension of the apnet library with a reproducible workflow.
Read the full abstract

Information Retrieval · Cross-listed from cs.LG

Defending against Model Extraction for GNNs with Model Reprogramming

Yan Wen, Zhenyi Wang and Heng Huang

Graph Neural Networks (GNNs) serve as the backbone for high-stakes applications in Machine-Learning-as-a-Service (MLaaS). Still, their black-box deployment exposes them to Model Extraction (ME) attacks, in which adversaries steal intellectual property by querying APIs. Existing defenses suffer from a critical ”Euclidean bias”: they transfer image-based strategies (e.g., random noise) to graphs, ignoring the complex topological dependencies between nodes, which often results in severe utility degradation. Passive methods like watermarking also fail to prevent theft in real time. To bridge this gap, we propose GraphRP (Graph Reprogramming Protection), a proactive defense framework that repurposes Model Reprogramming for security. Unlike static perturbations, GraphRP introduces a Structure-Aware Gating Mechanism driven by learnable topological prototypes. This creates a dynamic ”structural firewall” that selectively modulates the model's decision boundary: it preserves fidelity for benign queries residing on the training manifold, while maximizing the Fisher Information along the perturbation direction for adversarial queries. Under standard assumptions (bounded loss, optimal attacker, and local second-order approximation), we prove a lower bound on the attacker's estimation error that increases with the structural sensitivity of the reprogramming noise. Extensive experiments on both hard-label and soft-label ME attacks demonstrate that GraphRP significantly degrades attack effectiveness while preserving benign utility.
Read the full abstract

Information Retrieval

TRACES: A Benchmark for Epistemic Reliability in Scientific Reasoning by LLMs

Valentin Rodionov and Shamil Assylbekov

Large language models are being proposed as agents in scientific workflows, in domains where no downstream verifier exists. Such deployment assumes the model can distinguish reliable scientific literature from unreliable literature, a capability that has not yet been directly measured. Existing benchmarks evaluate factuality on questions with known answers; the failure mode we target here is different. We introduce a probe corpus of 42 retracted, fraudulent, and pseudoscientific papers, paired with a methodology for eliciting and scoring single-shot model engagement with each paper's framing. Each probe pairs a preamble extracted near-verbatim from the target paper with a scientifically plausible study-design request. The probes span five claim types: fabricated observation, pseudophysical mechanism, magical premise, legitimization bridge, and cargo-cult experiment. Two complementary scores measure whether a model rejects the flawed premise outright (IFR-a) and whether it recognizes the unreliability while still engaging (IFR-i). A depth score, the Engagement Depth Index (EDI), quantifies reproduction of paper- or field-specific withheld details. Across 30 models and 10 repeated runs, aggregate IFR-a is 0.93 $\pm$ 0.004 and aggregate IFR-i is 0.809 $\pm$ 0.009. Models engaged with untenable premises in 95% of all non-empty responses. Every evaluated model fails more than 71% of agentic probes, and 22 of 30 models fail more than 90% of the time. Rejections are concentrated on a small number of high-notoriety topics and specific probes, and disappear under matched-structure controls. These results are consistent with topic-keyed safety behavior rather than robust epistemic competence, and indicate an urgent need for guardrail infrastructure for scientific deployment of language models.
Read the full abstract

Information Retrieval · Cross-listed from cs.SI

Exploring the Social Life of Data: Finding Data You Can Trust

Penny R. Atkins and Manish Parashar

Artificial intelligence is changing the scale and tempo of scientific inquiry. Models can now search, integrate, and reason over data far beyond data repositories familiar to any individual researcher. Yet this expansion creates a prior problem: before a model can produce a trustworthy scientific result, it must locate data that are appropriate for the question, sufficiently reliable for the intended analysis, and accompanied by enough context to support responsible interpretation. As data becomes increasingly abundant, the challenge of finding data has been overcome by the challenge of finding data that you can trust. This paper explores how the social and empirical evidence that accumulates when data are used in research can be used, analogous to social trust networks, to determine fit for purpose and trust. Specifically, the paper explores data-usage graphs as a new layer of scientific data infrastructure. A data-usage graph connects datasets to the publications, people, institutions, topics, software, models, workflows, and other datasets through which they are produced and used. These connections reveal the social life of data: who has relied on a source, for which questions, in what combinations, with which methods, and with what observable impact. They can turn scattered traces of practice into data-usage descriptors that complement conventional metadata and support judgments of trust and fitness for purpose. The central claim is not that popularity establishes trust, but that this can be grown with appropriate contextual history. Usage evidence must therefore be combined with production quality, provenance, governance, semantic clarity, and community validation. The feasibility and value of data usage graphs is demonstrated by implementing the prototype data insights discovery service within the National Data Platform (NDP).
Read the full abstract

Information Retrieval · Cross-listed from cs.AI

Can Frontier LLMs Match Natively Multimodal Embeddings? A Comparison on Hard-Negative Text-to-Image Retrieval

Archan Dutta and Vyanktesh Kanungo

Multimodal retrieval and classification across different types of media, spanning text, images,video and audio, has traditionally relied on dual-encoder models that align visual and textual representations through contrastive learning. The March 2026 release of Gemini Embedding 2, Google's first natively multimodal embedding model to map text, images, video, audio, and documents into a single shared space, raises competition among multimodal retrieval systems. Simultaneously, frontier Large language models (LLMs) have also demonstrated strong visual understanding, raising the question of whether they can serve as effective zero-shot rankers. Our study provides the first direct comparison of native multimodal embeddings against LLM-based visual ranking on Flickr30k. We observe that GPT-4.1 and Claude Sonnet 4.6 perform on par with Gemini Embedding 2. Additionally, once embeddings are precomputed, multimodal embeddings are better suited for low-latency applications.
Read the full abstract

Information Retrieval

Are We Really Making Progress in Group Recommendation? Unmasking the Tie-Breaking Illusion

Song-Duo Ma and Pu-Jen Cheng

Recent group recommendation methods have reported strong improvements on standard benchmarks, but it remains unclear whether these gains always reflect genuine advances in modeling group preferences. In this paper, we show that several recent methods are affected by a systematic evaluation bias caused by the interaction between training-time score compression and evaluation-time deterministic tie-breaking. Specifically, an additional sigmoid transformation before the BPR objective can greatly increase tied top scores, making top-K metrics such as HR@K and NDCG@K highly sensitive to how ties are resolved. We revisit recent representative methods and their baselines on CAMRa2011 and Mafengwo under both group and user recommendation settings, and evaluate them with a tie-aware protocol that computes the exact expectation of HR@K and NDCG@K under uniform random tie-breaking. Our results show that many previously reported improvements shrink substantially under tie-aware evaluation, and the relative ranking of methods can change markedly. We further show that the additional sigmoid may act as implicit margin smoothing during optimization, and that temperature-scaled BPR can retain much of this benefit without inducing severe tie inflation. Overall, our findings highlight the importance of tie-aware evaluation for establishing reliable progress in group recommendation. The code is available at https://github.com/songduoma/TieAwareGroupRec.
Read the full abstract

Information Retrieval

Role of Personality in Conversational Information Seeking

Abdisalam Abukar, Junchen Fu, Chengli Zhai et al. (+1 more)

Large language models (LLMs) are increasingly used for information seeking, where users find, compare, and evaluate information through dialogue. In this role, the assistant does more than retrieve or generate content: it shapes how users articulate constraints, ask follow-up questions, verify claims, and decide when an answer is sufficient for action. Yet little is known about how user personality, assistant personality, and task context jointly influence these interactions. We examine personality as a controllable variable in conversational information seeking and study its effects on user behaviour and interaction quality. We conducted a controlled within-subject study in which assistant personality and task type were experimentally varied, while participant personality was measured using Big Five scores. Twenty-six participants each completed three information-seeking tasks under three assistant personality conditions: extraverted, conscientious, and neutral. Tasks covered exploratory travel planning, comparative smartphone shopping, and verification-sensitive health and diet information seeking. Data included conversation logs, behavioural traces, post-interaction questionnaires, an exit questionnaire, and Big Five measures. The assistant conditions were behaviourally distinct: the extraverted assistant produced longer turns, the conscientious assistant elicited higher user word share and more turns, and the neutral baseline fell between them. The strongest effect was a task-by-assistant interaction on trust and delegation, with preferred styles varying by task. No global winner emerged, but participants strongly preferred style choice or adaptation. These findings position assistant personality as a context-sensitive interactional design variable rather than a globally optimisable system property.
Read the full abstract

All authors Abdisalam Abukar, Junchen Fu, Chengli Zhai, Joemon M. Jose

Information Retrieval · Cross-listed from cs.CV

Multi-Level Evidence Aggregation for Robust Facial Phenotype Retrieval in Rare Genetic Disorder Prioritization

Alexander Hustinx, Carolin Kaffiné, Behnam Javanmardi et al. (+2 more)

AI-assisted facial phenotyping supports rare genetic disorder prioritization by retrieving visually similar diagnosed cases from facial image reference databases such as the GestaltMatcher Database (GMDB). Existing GestaltMatcher-based retrieval frameworks compare each test image with individual gallery images in a facial phenotype embedding space. However, this pointwise formulation does not fully exploit available evidence, because patients may have multiple images and disorders may be represented by multiple diagnosed gallery patients. We propose an inference-time multi-level evidence aggregation framework that improves facial phenotype retrieval without modifying the underlying GestaltMatcher-Arc encoder. The framework combines embedding-level patient aggregation of multiple images from the same individual, patient-weighted disorder centroids, and hybrid individual-centroid scoring to integrate test-patient observations, disorder-level gallery evidence, and local nearest-neighbor evidence. We evaluated the approach on GMDB v1.1.4 across disorders represented during training (GMDB-Freq), unseen disorders (GMDB-Rare), and multi-image patient subsets, using a unified gallery containing both GMDB-Freq and GMDB-Rare disorders. Multi-level evidence aggregation improved mean per-disorder top-$N$ retrieval accuracy across all evaluation subsets. Top-1 accuracy increased from 38.52% to 48.82% on GMDB-Freq and from 19.38% to 23.79% on GMDB-Rare. On multi-image subsets, top-1 accuracy increased from 46.12% to 60.94% on GMDB-Multi-Freq and from 18.54% to 26.71% on GMDB-Multi-Rare. These findings show that inference-time aggregation can improve next-generation facial phenotype retrieval without retraining the encoder, supporting a shift from isolated single-image matching toward multi-level aggregation of patient and disorder evidence for rare-disorder prioritization.
Read the full abstract

All authors Alexander Hustinx, Carolin Kaffiné, Behnam Javanmardi, Tzung-Chien Hsieh, Peter Krawitz

Information Retrieval

Self-Knowledge Retrieval Augmented Generation Framework for Patent Matching

Jian Zhang, Songlin Lei, Zhuohao Yang et al. (+6 more)

Patent retrieval and matching based on large language models (LLMs) play a vital role in intellectual property protection. However, due to the complex structure of patent documents, dense technical terminology, and multi-modal information, traditional methods struggle to accurately identify subtle differences between patents. Existing LLM-based patent matching approaches typically rely on domain-specific pretrained or instruction tuning, which often entail high manual labeling costs and catastrophic forgetting. While retrieval-augmented generation (RAG) methods introduce external knowledge they fail to fully leverage LLM's capability to automatically parse patents and mine deep semantic relationships. To address these limitations, this paper proposes a self-knowledge RAG framework that guides LLMs to autonomously extract key technical entities and construct hierarchical ontological structures from patent matching queries, thereby enabling query expansion and precise retrieval. The method integrates the FAISS retrieval with a generative matching mechanism, leveraging self-knowledge to enhance the model's understanding of patent innovations and significantly improve retrieval and matching accuracy. Experimental results demonstrate the outstanding performance of the proposed method on real-world patent datasets, validating its effectiveness and application potential.
Read the full abstract

All authors Jian Zhang, Songlin Lei, Zhuohao Yang, Bangli Liu, Ziwei Wang, Xufeng Weng, Gehan Amaratunga, Yu Lin, Hongwei Wang

Information Retrieval

Sona Technical Report

Sona Team: Alexandr Udeneev, Aleksei Krasilnikov, Alexey Nadtochiy et al. (+31 more)

We introduce Sona, a single-model generative recommender for Yandex Music. In an online A/B test, Sona replaced the entire production cascade, comprising more than 15 candidate generators followed by pre-ranking and ranking models that consume hundreds of features, including signals from large transformer models such as Argus and target-attention scorers, while significantly improving key engagement metrics. The architecture of Sona unifies candidate generation and ranking around a shared user representation. Its encoder transforms the user's chronological sequence of logged engagement events into hidden states consumed by both the autoregressive decoder and the Ranking Module. The next-token-prediction and distillation objectives jointly update the encoder, coupling generation and ranking through the same user state. Neither Sona nor its Teacher Ranker uses hand-engineered features; both operate on logged event fields and learned item representations. In the final Sona configuration, the larger teacher supplies ranking targets during training but is absent from serving, leaving the encoder, decoder, and Ranking Module as a single deployed model. We evaluate Sona in an online A/B experiment using live traffic from My Vibe on smart speakers, one of Yandex Music's largest recommendation surfaces. Relative to the production control, Sona produced statistically significant uplifts of 4.53% in Active Users, the primary metric, 6.30% in Total Listening Time, and 11.42% in Likes. These effects were incremental to improvements retained from preceding deployments. The Active Users uplift was 2.35 times the increment previously delivered by Argus, the strongest model deployed on this surface before Sona. These results show that a single jointly trained model can replace a mature multi-stage recommendation cascade while improving recommendation quality on live traffic.
Read the full abstract

All authors Sona Team: Alexandr Udeneev, Aleksei Krasilnikov, Alexey Nadtochiy, Andrey Semenov, Andrey Tsyrkunov, Anna Krivonos, Anna Lipkina, Artem Matveev, Daniil Burlakov, Daniil Leshchev, Daria Tikhonovich, Denis Burshtein, Ekaterina Dmitrieva, Eugene Krofto, Grigorii Khlystov, Ilya Murzin, Kirill Golovko, Ksenia Sycheva, Leonid Dmitriev, Mariia Rozaeva, Mariia Ulianova, Mikhail Sandul, Nikolai Savushkin, Oleg Sorokin, Roman Odobesku, Semyon Panenko, Sergei Liamaev, Sergei Makeev, Vadim Shilov, Veronika Ivanova, Viktor Yanush, Vladimir Baikalov, Vladislav Dodonov, Vladislav Tytskiy

Information Retrieval

TimeRoute: Time-Aware Modality Routing and Diffusion for Multi-Modal Recommendation

Pengyu Zhang, Yangqin Jiang, Klim Zaporojets et al. (+2 more)

Multi-modal recommenders fuse user-item interaction signals with item modalities such as text, images, and audio, but the usefulness of each drifts over time and at different rates. For example, around Valentine's Day, chocolate purchases become less driven by textual ingredient cues and more by visual packaging and ambient audio. This modality time-scale mismatch gives rise to two coupled challenges: (1) users with different temporal behavior profiles require different modality proportions, and (2) less relevant modalities are more likely to introduce outdated or misleading signals into the recommender. We address both challenges within a unified diffusion-based recommender, TimeRoute. A temporal-aware modal router maps each user's aggregated temporal profile to a personalized modality distribution, replacing the globally shared fusion weights used in prior work. The diffusion-based graph reconstructor is conditioned on the same profile through Feature-wise Linear Modulation (FiLM) with dual-stream long- and short-term denoising heads. This design captures both slowly and rapidly evolving temporal dynamics to suppress outdated modality edges before they enter the propagation graph. Experiments on TikTok, Amazon-Baby, and Amazon-Sports, averaged over 10 seeds, demonstrate consistent improvements over strong baselines across Recall@K, Precision@K, and NDCG@K, reaching up to 9.8% (P@20 on Amazon-Baby). Controlled attribution studies further show that these gains require both the proposed mechanisms and temporal input: naively granting the backbone the same temporal profile yields no benefit, and feeding the router random noise performs no better than removing the router entirely. Code is available at https://anonymous.4open.science/r/TimeRoute.
Read the full abstract

All authors Pengyu Zhang, Yangqin Jiang, Klim Zaporojets, Congfeng Cao, Paul Groth

Information Retrieval

Deciding When to Rely on Visual Information: Gated Multimodal Fusion in Sequential Recommendation

Natalija Glisovic, Danica Kragic and Martin Tegner

Multimodal sequential recommender systems commonly fuse visual and collaborative signals uniformly, treating visual features as generically informative regardless of item or user context. We argue that visual utility, defined as the contribution of visual signals to recommendation quality, is a latent contextual variable that depends on both the item and the user's interaction history rather than a fixed item property. To model this variability, we introduce VisGate, a framework that makes adaptive item-level fusion decisions conditioned on item embeddings and the user's current sequence context. Visual representations are learned through a contrastive objective over sequential co-occurrence patterns, preserving complementarity with collaborative embeddings rather than aligning them into a shared space. Beyond achieving competitive recommendation performance, VisGate's learned gate serves as a measurement tool for understanding when and why visual information is beneficial. Our analyses show that visual utility varies across items, increases under interaction sparsity when collaborative signals are weak, and correlates with visual distinctiveness in semantically meaningful ways. Together, these findings highlight the importance of both fine-grained fusion and modality complementarity, while demonstrating that item-level visual utility can be estimated and interpreted through learned gating behaviour.
Read the full abstract

Information Retrieval · Cross-listed from cs.CL

Leveraging Human Reading Behavior for Keyphrase Extraction: A Webcam-based Eye-tracking Corpus

Chengzhi Zhang, Xinyi Yan and Wenqi Yu

Purpose: Keyphrases are statistically and semantically important textual units that can also attract readers' attention during comprehension. However, existing keyphrase extraction (KPE) studies mainly focus on improving textual representation while largely overlooking human reading behavior. This study examines whether lightweight webcam-based eye-tracking features can improve KPE from Chinese academic abstracts in Library and Information Science (LIS). Methodology: To address the limited availability of eye-tracking data for Chinese academic reading, we developed a lightweight webcam-based data collection platform using the open-source SearchGazer library and constructed the Chinese LIS Eye-Tracking Corpus (CLIS-ET). Three character-level eye-tracking features, first fixation duration (FFD), fixation number (FN), and total fixation duration (TFD), were incorporated into KPE models to evaluate their effects on extraction performance. Findings: Eye-tracking features consistently improved KPE performance. The combination of FN and TFD achieved the best results on the Att-BiLSTM+CRF model, indicating that readers' fixation behavior provides useful signals for identifying keyphrases in academic abstracts. Originality/value: This study introduces a cost-effective webcam-based eye-tracking approach for KPE and presents CLIS-ET, a Chinese academic eye-tracking corpus containing FFD, FN, and TFD features. The results demonstrate the value of incorporating human reading behavior into keyphrase extraction. Dataset and code: https://github.com/yan-xinyi/ET_AKE and https://github.com/yan-xinyi/Reading_ET_System.
Read the full abstract

Information Retrieval

ENTLORE: A Graph-Grounded Benchmark for Latent Organizational Reasoning in Enterprise Question Answering

Akrin Zheng, Alexander Wu and Alaia Liu

Enterprise question answering is framed as retrieving internal documents and generating grounded answers. Routine enterprise records, however, are work by-products in which required organizational relations remain implicit across heterogeneous sources. Existing benchmarks provide realistic multi-source evidence, but often materialize a predefined answer path and therefore test the composition of stated facts rather than recovery of a target relation absent from the corpus. We call the latter capability latent organizational reasoning. We introduce ENTLORE, a graph-grounded benchmark construction framework that reconstructs an audited enterprise world from routine documents, authoritative organizational tables, and operational records. Versioned organizational conventions certify derived relations in a truth graph, enabling complete golden answers and proof certificates. The aligned anonymized release exposes only the document corpus while withholding private structure and target relations. ENTLORE contains 2,341 documents from three source types and 907 questions spanning explicit lookup, cross-source composition, and latent organizational reasoning, evaluated across 56 model and access configurations. Structuring the released world as an induced entity graph or navigable knowledge base gives the strongest deployable results. Yet supplying gold documents still leaves 30.4% of latent questions unanswered, versus 12.6% and 6.2% for explicit and compositional questions. Enterprise QA therefore depends not only on document recall, but also on whether implicit organizational relations become usable. The benchmark, data, and code are publicly available at https://github.com/scitix/entlore .
Read the full abstract

Information Retrieval

DistilVDR: A Compact End-to-End Visual Document Retriever via Dual-Student Distillation

Zhuchenyang Liu, Ziyi Wang, Yao Zhang et al. (+1 more)

Visual document retrieval (VDR) is dominated by multi-billion-parameter models that are slow to index at full corpus scale and expensive to serve. Prior compression routes either train a smaller multi-vector encoder from scratch or distil only the query side; neither yields a compact single-vector retriever end-to-end. We present DistilVDR, a 524M end-to-end VDR system distilled bilaterally from a single 8B vision-language teacher under a pointwise cosine alignment loss. All supervision comes from the frozen teacher's embedding space, which was itself trained with relevance supervision, so the student objective needs no relevance labels, negative sampling, or contrastive term. We match VDR's text-query and image-document input asymmetry with an asymmetric encoder-only student that concentrates visual capacity on the document side and keeps the query side at 70M parameters. We release two variants that share the same encoders and training and differ only in the document encoder's visual-tile budget: DistilVDR-HiRes attains 61.74 average NDCG@5 on ViDoRe v1+v2+v3 (86.9% of the 8B teacher) and leads every reproduced sub-1B baseline on the high-resolution-sensitive v3 benchmark, while DistilVDR-Fast attains 59.98 at a 3 times smaller visual-token budget. Both variants store one million documents in a 15.6 times smaller index than the strongest sub-1B multi-vector baseline and index the corpus an order of magnitude faster. The code is available at https://github.com/Ryenhails/NanoVDR.
Read the full abstract

All authors Zhuchenyang Liu, Ziyi Wang, Yao Zhang, Yu Xiao

Information Retrieval

Multi Interests for Joint Search-Recommendation Modeling

Xiangchen Pan, Wei Wei, Huakang Niu et al. (+1 more)

Search and recommendation are crucial for understanding user preferences. More and more studies are attempting to jointly model search behavior and recommendation behavior, by integrating user active search and passive recommendation behavior data to better mine user preferences. However, although existing cross-domain unified modeling frameworks can effectively compensate for the differences in behavior between domains, they overlook the expression of interests in different scenarios under mixed sequences. In this study, we propose a multi-interest-based mixed sequential modeling framework MIJSR, which performs multi-interest mining and adaptive integration on search recommendation mixed sequences from both structural and semantic perspectives. Specifically, our model can be roughly divided into three modules: cross-domain behavior fusion, multi-interest mining, and multi-task prediction. Firstly, we align the representations of query and item through contrastive learning training. Then, we extract the multi interests of the mixed behavior sequence from both structural and semantic perspectives. Structurally, we extract search interests, recommendation interests, and cross interests through subsequence partitioning and mask settings; In terms of semantics, we use the semantic information of queries for clustering and perform semantic segmentation on mixed sequences to construct semantic multi interests. Finally, the adaptive fusion of multiple interests is combined with other side information to use a progressive layered extraction model for multi-task prediction. Extensive experiments on two open-source datasets have shown that our model can further enhance its accuracy in search and recommendation by extracting users' multi interests at a fine-grained level. Codes are available at https://github.com/pxcstart/MIJSR.
Read the full abstract

All authors Xiangchen Pan, Wei Wei, Huakang Niu, Zhicong Cheng

Information Retrieval

When Do Anchor-Based Pointwise LLM Rerankers Help? Retriever Quality, Statistical Scope, and Anchor Design

Utshab Kumar Ghosh and Shubham Chatterjee

Anchor-based pointwise LLM reranking scores each candidate against a shared reference passage to recover cross-document context at pointwise cost. We study when this actually helps, using GCCP/PAGC as a representative method. Our study is reproduction-first. We use reproduction as a starting point for a controlled component-level stress test of anchor-based pointwise reranking. Our initial reimplementation, based only on the paper text, achieves 0.24 nDCG@10 instead of the reported 0.66, revealing that several undocumented implementation details are necessary to reproduce the method. After identifying and recovering eight such details, we reproduce the reported results within 1.6% and use the validated implementation for controlled analysis. We find that the core contrastive scoring idea is robust under rigorous statistical correction. However, two design choices held fixed in the original paper are less reliable. First, we find that combining the contrastive score with the standard pointwise relevance score helps when the first-stage retriever is BM25, but gives little or no benefit when the first-stage retriever is a stronger dense model such as E5. Second, the paper's more complex method for constructing the anchor is unnecessary. A much simpler anchor, built by interleaving the top-ranked sentences, matches or outperforms it across datasets. These findings are consistent across different LLM backbones, including a 4-bit quantized 72B model. Overall, anchor-based pointwise reranking is effective, but its gains come mainly from contrastive scoring rather than from the more complex aggregation and anchor-construction choices, and they appear under narrower conditions than the original evaluation suggests.
Read the full abstract

Information Retrieval

Towards Efficient Reasoning in LLM-Based Recommender Systems via Model Merging

Linh Dieu Le, Tong Chen, Shazia Sadiq et al. (+3 more)

Large language model-based recommender systems are increasingly adopting slow-thinking models that generate step-by-step reasoning before making predictions, often achieving higher accuracy than fast-thinking models that predict directly. However, their reasoning traces are often unnecessarily verbose, increasing inference costs without commensurate accuracy gains. Existing training-based approaches to reasoning compression often incur substantial adaptation costs, while inference-time methods are brittle and difficult to scale. These limitations motivate model merging as a promising training-free direction for transferring specialised behaviours between models in a shared parameter space. In particular, merging a slow-thinking model with a fast-thinking counterpart provides a natural mechanism for balancing recommendation accuracy and reasoning conciseness. To this end, we propose, to our knowledge, the first model merging framework for reasoning compression in recommender systems. Unlike conventional merging methods that apply uniform merge coefficients across model components, our method performs fine-grained merging at the level of individual attention heads, capturing heterogeneous patterns in recommendation reasoning. Each attention head is assigned a distinct merge coefficient according to its contribution to critical reasoning evidence and its sensitivity to parameter change, enabling selective injection of the concise behaviour of the fast-thinking model into the slow-thinking model and reducing reasoning verbosity without compromising recommendation quality. Experiments on three benchmark datasets show that our method reduces reasoning length by up to 24.3% while outperforming competitive model merging baselines in maintaining recommendation accuracy. The code is available at https://github.com/linhledieu/REAM.
Read the full abstract

All authors Linh Dieu Le, Tong Chen, Shazia Sadiq, Hongzhi Yin, Ming Jin, Junliang Yu

Information Retrieval · Cross-listed from cs.LG

RAISE: Diagnosing Acquisition Collapse in Costly LLM Signals

Ying Yuan, Yu Wang, Yize Cheng et al. (+1 more)

Large language models (LLMs) are increasingly used as costly, on-demand components in real systems, but calling them indiscriminately can waste substantial compute, latency, and serving budget. The key deployment question is therefore not only whether an LLM helps on average, but when it is worth calling. We identify a common failure mode, which we call acquisition collapse: an LLM signal can appear useful in aggregate or post hoc, yet still provide too little before-call information to support reliable selective use. We introduce RAISE (Reward-SNR Actionability in Signal Evaluation), a pre-routing diagnostic framework for testing whether available evidence supports selective use before committing to a routing strategy. We instantiate RAISE with Structured Hypothesis Embeddings (SHE), a frozen-LLM intent signal for recommendation using one LLM call per user, and evaluate it through controlled, retrospective, and fresh-cohort studies and a prospective offline pilot whose audit decisions are frozen before independent outcomes are revealed. Across these settings, predictable incremental benefit, not average lift alone, distinguishes settings with recoverable selective value; deployment additionally depends on cost and operational constraints. Seemingly strong oracle or subgroup gains can disappear under independent evaluation. More broadly, RAISE reframes costly inference as an information-acquisition problem: before paying for an expensive model, tool, sensor, or measurement, first test whether its value is predictable at decision time. This principle motivates cost-aware acquisition in settings ranging from agent tool use and stronger-model consultation to robotic sensing and clinical decision pipelines.
Read the full abstract

All authors Ying Yuan, Yu Wang, Yize Cheng, Xuyang Wu

Information Retrieval

Post-Calibration Reliability Reranking of Relevance Decisions via Label-wise Monotone Projection

Inwoo Tae and Yongjae Lee

Web search, product search, and question-answering retrieval systems often assign a relevance label and confidence score to each query-candidate pair. The relevance label describes how well a page, product, or passage matches the query, while the confidence often guides downstream use or fallback decisions. Post-hoc calibration is therefore needed because misaligned confidence can make systems over-trust wrong predictions or unnecessarily defer correct ones. However, calibration mainly aligns confidence with average correctness, and does not remove predicted-label-dependent reliability differences that remain within the same calibrated confidence level. We address this gap with Label-wise Monotone Reliability Projection (MRP), which learns label-wise monotone functions that map calibrated confidence to correctness reliability while preserving the original predicted labels and class probabilities. The resulting reliability score reranks fixed predictions according to residual risk. Across six information access relevance datasets and multiple post-hoc calibrators, MRP improves reliability reranking and average fallback utility while preserving full-coverage accuracy and ECE. Structural ablations show that the main gains come from label-wise residual reliability rather than from global confidence remapping. We further analyze when MRP reliability scores can be embedded back into top-label probability geometry, showing that this projection is useful as a compatibility analysis but is distinct from the main reliability-reranking objective. The implementation will be made publicly available.
Read the full abstract

Information Retrieval

Persona Conditioning as an Assessor-Sensitivity Probe for LLM-Based IR Evaluation

Samaneh Mohtadi, Pietro Bernardelle, Joel Mackenzie et al. (+1 more)

Large language models (LLMs) are increasingly used as relevance assessors in information retrieval (IR) evaluation, raising questions about how assessor framing affects judgment reliability and downstream system comparison. We study persona conditioning as a diagnostic mechanism for exposing LLM assessor sensitivity. Using task-oriented personas drawn from two complementary sources (PersonaHub and NVIDIA Nemotron-Personas-USA), we instantiate five assessor roles emphasizing intent interpretation, domain expertise, contrastive judgment, evidence verification, and global search-quality assessment, compared with a standard UMBRELA baseline. Across six LLM backbones on TREC DL20 and RAG24, our analyses reveal structured rather than uniform assessor sensitivity. Judgments usually remain close to the baseline while shifting assessment strictness, evidential threshold, or interpretation emphasis rather than producing widespread relevance reversals. At the system level, high-capacity models preserve system-ranking agreement, while smaller models amplify persona-induced instability. Local rank-displacement analysis shows sensitivity concentrates on particular retrieval systems and system types, especially neural ranking/reranking systems on DL20 and RAG-oriented pipelines on RAG24. Persona source matters less than assessor role and model capacity. These findings position persona-conditioned judging as a controlled sensitivity probe for stress-testing LLM-based IR evaluation pipelines and identifying systems whose evaluation outcomes are sensitive to assessor framing.
Read the full abstract

All authors Samaneh Mohtadi, Pietro Bernardelle, Joel Mackenzie, Gianluca Demartini

Monday 10 August

Information Retrieval

When Concealed Links Cannot Be Recovered: A Structural Identifiability Bound and Evaluation Pitfalls in Offshore Leak Networks

Joseph Bingham

Leaks such as the Panama and Paradise Papers expose large networks of offshore entities, and they invite an obvious question for network learning. Can the relations these structures are built to hide—above all, who beneficially owns what—be recovered from the public part of the leak by link prediction? We argue that the answer is mostly no, and that the analyses which suggest otherwise are measuring the wrong thing. Our main result is a distribution-free identifiability bound. For any recovery rule that respects graph isomorphism, and that covers every topological link-prediction score together with every message-passing graph neural network, a concealed endpoint left isolated in the observed graph is interchangeable with its structural twins, so its hidden edge cannot be recovered above chance. Isolation is only the sharpest case. In general the ceiling on recovery is set by the size of a node's structural-indistinguishability class, for which the Weisfeiler–Leman colour class is a computable stand-in, and node degree is at best a loose proxy. On the full ICIJ Offshore Leaks graph (814K entities and 84K labelled beneficial-owner edges) a classifier-free, degree-controlled probe reproduces an exact $0.5$ floor for isolated owners, who make up $20.6\%$ of all owners, and a trained graph neural network lands on the same floor. Recovery climbs only as structural distinctiveness grows, and the floor reappears in every one of the five leaks. Along the way we document five evaluation traps. Each one makes a bound that cannot be beaten look beaten, and we give a short rule that avoids them. The practical upshot is to redirect effort from the hidden principal, which is close to unrecoverable, toward the machinery of concealment, and to spell out why fusing external data helps far less than one would hope.
Read the full abstract

Information Retrieval · Cross-listed from cs.SI

Graph Neural Team Recommendation: An Integrated Approach

Md Jamil Ahmed, Mahdis Saeedi and Hossein Fani

Team recommendation aims to select an optimal subset of experts who can form an almost surely successful collaborative team for a given set of required skills. State-of-the-art methods are neural multi-label classifiers that transfer dense vector representations of skills into a sparse occurrence vector representing the optimal subset of experts. Such methods, however, overlook experts' relational and structural information encoded in the expert collaboration graph and, thus, fall short of capturing complex inter-dependencies among experts and their associated skills within teams. Moreover, the skills' dense vectors are pretrained disjointly and independently of the underlying neural classifier, hence, preventing end-to-end optimization. In this paper, we propose to reformulate the team recommendation problem into end-to-end link predictions in the expert collaboration graph to consume multi-hop intra-team and cross-team collaborations among experts while eschewing the unnecessary complexities of the disjoint two-phase training procedure. Our experiments on two large-scale datasets from various domains with distinct distributions of skills in teams demonstrate the superiority of the end-to-end approach and establish a new state of the art. Our code is available at https://github.com/fani-lab/OpeNTF.
Read the full abstract

Information Retrieval

Neural Tree Collaborative Filtering: Rethinking Graph Collaborative Filtering as Tree Collaborative Filtering with Curvature-Aware Propagation Depth

Jinfeng Xu, Zheyu Chen, Ziyue Peng et al. (+5 more)

Graph Collaborative Filtering (GCF) has become the dominant paradigm in modern recommender systems by modeling user-item interactions as a bipartite graph and propagating embeddings through a fixed number of message-passing layers. However, applying a uniform propagation depth to every node ignores a fundamental property of real interaction graphs: nodes differ substantially in their local connectivity, so peripheral nodes quickly suffer from over-smoothing while hub-like nodes remain under-explored beyond their immediate neighborhood. In this paper, we revisit GCF from a tree-structured perspective and propose Neural Tree Collaborative Filtering (NTCF), a framework that re-interprets each node's local neighborhood as a rooted tree and assigns a node-specific propagation depth based on a closed-form local-degree-imbalance score that serves as a discrete Ricci-curvature proxy. We provide a theoretical analysis showing that (i) NTCF strictly generalizes NGCF, degenerating to NGCF when all curvature-induced depth adjustments vanish (a lower bound on its representation power), and (ii) the curvature-aware schedule retains strictly more discriminative information at deep layers on positively-curved (peripheral) nodes than uniform-depth propagation. NTCF can achieve higher performance than most widely used GCF backbone models and can be integrated into existing advanced self-supervised models as a backbone, replacing their original backbone to achieve enhanced performance. Extensive experiments on three public datasets demonstrate the superiority of NTCF.
Read the full abstract

All authors Jinfeng Xu, Zheyu Chen, Ziyue Peng, Shuo Yang, Jinze Li, Wenhao Yuan, Jian Chen, Edith C. H. Ngai

Information Retrieval

GenRec: An LLM-Backed Recommendation Ranker at Netflix

Ying Li, Shradha Sehgal, Arjun Rao et al. (+3 more)

Large language models (LLMs) are reshaping recommender systems by enabling richer modeling of users, content, and context directly in natural language. At Netflix, we are exploring this direction through GenRec, an LLM-backed recommendation ranker built on top of an in-house foundational LLM. GenRec follows a two-phase framework: Phase 1 adapts an open-source LLM to Netflix data, developing deep understanding of the catalog and member behavior while balancing capabilities such as content understanding and instruction following. Phase 2 post-trains this foundation model with recommendation-ranking specific data, labels, and reward signals, aiming to align the ranker with business requirements and long-term member satisfaction. This paper focuses on Phase 2 and the transition from a traditional discriminative ranker with thousands of engineered features to an LLM-backed ranker driven by verbalized user histories and context. We describe our design for input verbalization and context engineering, post-training data construction, reward integration, model architecture, and a cost-constrained serving design based on a prefill-only inference approach. We report results from a large-scale A/B test comparing GenRec against the current production ranker model, where we show that a GenRec model trained with substantially fewer Phase-2 labeled training examples and input signals can achieve statistically significant gains in offline and online metrics. We discuss how LLM-backed recommenders could shift the recommendation paradigm: from feature engineering to context engineering, and from bespoke architectures to shared foundation backbones. We also outline practical lessons for serving such systems under real-world resource constraints.
Read the full abstract

All authors Ying Li, Shradha Sehgal, Arjun Rao, Rein Houthooft, Yaochen Zhu, Ashish Rastogi

Information Retrieval

DualSpectralCF: Training-Free Sign-Aware Spectral Collaborative Filtering

Guanqun Yang, Tong Qi and Xiaoxue Han

Real-world recommendation platforms routinely collect explicit negative feedback such as 1-star reviews, hate-button clicks, distrust between users, and very-low watch-ratio videos. Learned sign-aware recommenders exploit this signal for clear accuracy gains, but only at the cost of gradient-based training. In parallel, a line of training-free spectral collaborative filtering methods matches or beats learned graph recommenders at a fraction of the cost, yet operates on positive interactions alone. We bridge these two lines with DualSpectralCF, a training-free framework of two components that attach to any spectral backbone of the form $\hat{\mathbf{r}}_u = F(\mathbf{M}) \mathbf{r}_u$: a signed input signal $\mathbf{r}_u^{\pm}$ that encodes the user's explicit dislikes, and a signed item-item operator $\mathbf{M}^{\pm}$ that blends like-together and dislike-together similarity. The framework is backbone-agnostic and adds just two scalar hyperparameters. We instantiate DualSpectralCF on ChebyCF, GF-CF, and Turbo-CF, and evaluate on five sign-aware benchmarks: every instance matches or beats its unsigned backbone on all 5 datasets, with Recall@20 lifts up to +32.6% with backbone-specific $(\gamma, \kappa)$ tuning and +1.9% to +16.0% for DualSpectralCF-Cheby at the fixed default $(\gamma = -0.5, \kappa = 0.1)$, and the family runs 7.7 to 155.3$\times$ faster than SIGformer while reaching 70.7% to 90.7% of its accuracy. Sign-awareness helps most for cold-start users, with up to +29.2% Recall@20 on Epinions users with 1 to 5 training items.
Read the full abstract