SEARCH · Searcharxiv

Search Searcharxiv

Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 271 records · Page 15Linked to original sources

arXiv2026

Agents as Knowledge Integrator and Utilizer in Multimodal Recommendation

Online platforms increasingly rely on multimodal recommender systems to rank products, media, and other Web content. Existing methods usually inject visual and textual features into item representations or build homogeneous graphs from modality-level similarity, but the resulting signals can remain misaligned with the recommendation objective. We study this semantic gap from a knowledge-integration perspective: multimodal content should be interpreted together with user behavior before it is used to construct recommendation graphs or adjust rankings. We propose AgentMMRec, an agent-based multimodal recommendation framework with two coordinated roles. The Integrator Agent infers behavior- and multimodal-aware user preferences and item properties from training interactions and item content, then stores them in a reusable knowledge memory. The Utilizer Agent consumes this memory to refine modality-specific item-item graphs, construct behavior-aware homogeneous graphs, and rerank candidate lists under a frozen evaluation-time memory. This design differs from direct LLM feature augmentation and pure LLM reranking because the generated knowledge is first converted into graph structure and model representations before recommendation. Experiments on three Amazon multimodal recommendation datasets show that AgentMMRec consistently improves Recall and NDCG over recent multimodal baselines, remains effective under sparsity and item cold-start settings, and can transfer its constructed knowledge to existing backbones.

cs.IR↗

arXiv2026

Content Exploration Beyond the Feed: Creator Supply and the Shared Corpus

Industrial recommenders give new content initial views through budgeted exploration, then use early performance to decide further delivery. On many short-video platforms, exploration is the primary way new videos reach viewers. Viewer-side tests measure consumption; the published budget objectives we review omit creator response. We analyze four experiments on a major short-video platform. An eight-month creator ablation finds production exploration raises videos posted per creator by 8.55% and creators posting at least once by 7.10% relative to a minimal floor. A budget-matched reallocation raises creator participation with no detectable short-run viewer-side change. A year-long viewer ablation finds 1.74% more video views but 2.13% less view time. A delivered view creates immediate feed value, can trigger organic take-up, and can induce creator supply. Take-up and supply replenish a shared corpus, creating two measurement limits. Viewer-side A/B tests cancel the corpus effect when both arms consume the same corpus. Giving each arm its own corpus avoids cancellation, but turnover still controls the horizon. If the corpus turns over at rate w per posting cycle, a t-cycle experiment expresses at most wt of the eventual corpus effect. More users reduce noise but do not speed turnover. Before the corpus path visibly bends, data cannot distinguish a modest fast effect from an arbitrarily large slow one, so a valid confidence interval may lack a finite upper endpoint. As predicted, the three-week co-diverted experiment cannot determine the sign of the eventual corpus effect. Within the window, it identifies the direct feed effect, and an exploratory cohort analysis detects organic lift after exploration ends. The experiments establish a positive creator response, measure the gross corpus flow visible within three weeks, and show the design and duration needed to identify total value.

cs.IR↗

arXiv2026

SilentProbe: Measuring Silent Failure in Production APIs Used as Agent Tools

An LLM agent calling a production API cannot distinguish a query that matched nothing from a query the server did not understand. Both return HTTP 200 with a parsable body, no exception to catch and no field to branch on. We ask what predicts which one occurred, and what it does to the agent. Auditing 721,320 parameters across 2,501 independently published OpenAPI documents, we find that 7.5% declare an enumeration and 15.2% declare any machine-checkable constraint at all, while 40.1% of documents state at least one constraint in prose that their schema does not encode. Executing 219 schema-derived perturbations against live commercial endpoints from 27 vendors, reached through a single aggregation layer (Monid) that publishes a schema and returns a run identifier for every call, we find that constraint form, not vendor identity, predicts honesty: machine-checkable constraints yielded an honest error in 111 of 111 cases, prose-only constraints failed silently in 44 of 61 (p = 2e-13). Twelve models across eight families then met these endpoints on ordinary tasks. A vocabulary that the description merely exemplifies was missed by every model on 88 of 88 attempts, while vocabularies written out in full were used correctly 88 to 91% of the time. Running the full agent loop, models detected the resulting silent failure in 12% of cases, repaired it in 0%, asserted a false negative to the user in 41%, and invented a figure in 12%. Promoting the vocabulary into the schema removes the failure, from 88 of 88 to 0 of 89. The fix is one line of schema rather than a better model. Code, schemas, perturbation sets, agent transcripts and per-call run identifiers are released at https://github.com/Jasper0122/silentprobe.

cs.IR↗

arXiv2026

On the Consistency and Performance of the Iterative Bayesian Update

In many situations, estimating the distribution of users' data concerning certain attributes is important. To facilitate this estimation while safeguarding users' privacy, the local privacy model is commonly employed, in which each user applies a local protection mechanism to release a noisy version of their original data to the data collector. The original distribution is then estimated using methods such as Matrix Inversion (INV), RAPPOR's estimator, and iterative Bayesian update (IBU). In this article, we experimentally demonstrate that IBU significantly outperforms the other methods when user data is protected through metric privacy mechanisms. We also explain the mathematical reason for the suboptimal performance of INV under those metric privacy mechanisms. Conversely, IBU exhibits performance similar to INV under typical mechanisms of local differential privacy, specifically the k-RR and RAPPOR. In addition, we investigate IBU's consistency, which is a crucial property as it means that the estimate converges to the true distribution as the number of data points increases. In the literature, IBU is claimed to be consistent, but there is no proof for this claim. We provide a formal proof of consistency leveraging the fact that IBU is a maximum likelihood estimator. Finally, we examine scenarios involving an infinite alphabet for sensitive data and propose a method allowing IBU to operate effectively in these situations.

cs.CR↗

arXiv2026

CASTLE: Contrastive and Seed-Guided Training for Cold-Start Natural Language Search

Deploying natural language search systems presents a critical cold-start challenge: no real user queries to learn linguistic patterns, and no relevance labels to train ranking models. We present CASTLE (Contrastive And Seed-guided Training for natural Language sEarch), an LLM-based framework for generating synthetic queries and relevance labels from structured catalog data, powering Airbnb's natural language search across its full lifecycle. CASTLE makes three contributions. First, we generate realistic queries by combining structure-guided prompting with seed queries from user research, using template, few-shot, and attribute-grounded prompt variants together with explicit variety mechanisms to prevent query collapse. Second, we produce relevance labels by construction via contrastive listing pairs derived from booking sessions, achieving near-zero false positives without LLM judgment. Third, CASTLE's structured input design is flexible: incorporating richer signals such as guest reviews and photo captions alongside listing attributes enables generation of niche, long-tail queries that reflect subjective user preferences (e.g., "cozy cabin with fireplace") beyond what catalog attributes alone can express. Compared against InPars-style, Promptagator, and contrastive-only baselines, CASTLE achieves KL 1.01 vs. real users -- a 9.2x improvement over the best baseline (9.33) -- and the lowest attribute-type KL divergence (0.08), outperforming even survey seed queries (0.09). A human evaluation on 200 sampled triplets confirms label quality: annotators agree with CASTLE labels at 91-93%. We deploy production pipelines generating synthetic examples daily for embedding-based retrieval and ranking evaluation. Synthetic data remains valuable beyond cold-start: it targets tail queries underrepresented in organic traffic and extends naturally to multi-turn conversational search.

cs.IR↗

arXiv2026

Context-Aware Interpretable Representations for Retrieval and Graph Convolutional Network Classification

The advances in visual information modeling and representation during the last decades are remarkable, mainly supported by Convolutional Neural Networks, Transformer-based, and Foundation Models. Despite this progress, critical challenges regarding the nature of similarity assessment and model transparency have been neglected. A primary concern is the Geometric Gap, where traditional pairwise measures fail to capture the intrinsic geometry of the dataset manifold. Furthermore, the Interpretability Gap persists, as representations often lack alignment with human cognition. Therefore, how to provide interpretability to representations while maintaining low dimensionality and high effectiveness in downstream tasks remains an open challenge. In this paper, we propose a novel unsupervised framework that integrates Manifold Learning strategies with Rank-based Interpretable Graph Embeddings. Our approach effectively bridges these gaps by first characterizing the contextual information of the dataset through manifold analysis and subsequently generating sparse, self-explainable embeddings. The proposed approach employs a flexible formulation, allowing different Manifold Learning and Representation Learning strategies. Extensive experimental evaluation across diverse datasets and features demonstrates that our Context-Aware representations not only provide intrinsic interpretability and dimensionality reduction but also maintain or enhance effectiveness in downstream tasks, specifically in image retrieval and semi-supervised classification using Graph Convolutional Networks (GCNs).

cs.LG↗

arXiv2026

An Empirical Study on Zero-Data Bootstrapping for Conversational Recommender Systems

Conversational Recommender Systems (CRS) typically require domain-specific dialogue data, which is costly, scarce, and often unavailable in new domains. We conduct a systematic empirical study of zero-data CRS bootstrapping: generating synthetic conversational supervision from non-conversational signals---item reviews, metadata, and user-item interactions---without any in-domain dialogue corpus. We compare two information-theoretic selection strategies, Jensen-Shannon diversity and Fisher information, across domain signals, model architectures, datasets, and fine-tuning paradigms. Our results show that domain-grounded synthetic data consistently outperforms zero-shot prompting and naive synthetic baselines; active selection improves data efficiency over random sampling; metadata and collaborative filtering signals each improve selection quality; and, in low-resource settings, synthetic data can outperform scarce real dialogues while further complementing them. These findings establish non-conversational domain signals as a viable path toward building CRS without conversational training data. The code is available at https://anonymous.4open.science/r/zero_data_crs/ .

cs.IR↗

arXiv2026

GPU-Native Approximate Nearest Neighbor Search with IVF-RaBitQ: Fast Index Build and Search

Approximate nearest neighbor search (ANNS) on GPUs is gaining increasing popularity for modern retrieval and recommendation workloads that operate over massive high-dimensional vectors. Graph-based indexes deliver high recall and throughput but incur heavy build-time and storage costs. In contrast, cluster-based methods build and scale efficiently yet often need many probes for high recall, straining memory bandwidth and compute. Aiming to simultaneously achieve fast index build, high-throughput search, high recall, and low storage requirement for GPUs, we present IVF-RaBitQ (GPU), a GPU-native ANNS solution that integrates the cluster-based method IVF with RaBitQ quantization into an efficient GPU index build/search pipeline. Specifically, for index build, we develop a scalable GPU-native RaBitQ quantization method that enables fast and accurate low-bit encoding at scale. For search, we develop GPU-native distance computation schemes for RaBitQ codes and a fused search kernel to achieve high throughput with high recall. With IVF-RaBitQ implemented and integrated into the NVIDIA cuVS Library, experiments on cuVS Bench across multiple datasets show that IVF-RaBitQ offers a strong performance frontier in recall, throughput, index build time, and storage footprint. For Recall approximately equal 0.95, IVF-RaBitQ achieves 3.0x higher QPS than the state-of-the-art graph-based method CAGRA, while also constructing indices 14.7x faster on average. Compared to the cluster-based method IVF-PQ, IVF-RaBitQ delivers on average over 4.5x higher throughput while avoiding accessing the raw vectors for reranking.

cs.DB↗

arXiv2026

Building Better Encoder-only Cross-Encoders: A Controlled Study of Training Strategies for Neural Re-ranking

Cross-encoders fine-tuned from Transformer backbones remain the standard for second-stage re-ranking, and recent knowledge-distillation strategies have closed much of the gap with LLM re-rankers. However, these strategies have not been compared under controlled conditions. In particular, it remains unclear how distillation from LLM rankers compares to distillation from strong cross-encoder teachers, or to purely supervised objectives. It is also unclear how much newer backbones (RoBERTa, ELECTRA, DeBERTaV3, ModernBERT) contribute compared to the original BERT. We run 162 controlled training runs (9 backbones x 6 objectives x 3 seeds), spanning pointwise, pairwise, and listwise losses with both human labels and two distillation signals, and evaluate on TREC-DL, MSMARCO dev, BEIR, LoTTE, and Robust04. We find that objectives emphasizing relative comparisons - pairwise MarginMSE and listwise InfoNCE - consistently outperform alternative objectives, including more complex listwise LLM distillation, across all backbones, and switching objective yields gains comparable to moving up one backbone size tier. A controlled disentanglement further shows that, once the negative-sampling pool is matched, even a simple pairwise Hinge loss with ColBERTv2 hard negatives matches - and on out of domain beats - listwise LLM distillation, indicating that the quality of the negatives is at least as important as the choice of loss. We release all 162 trained models on HuggingFace (https://huggingface.co/collections/xpmir/reproducing-cross-encoders) and a unified training codebase. (https://github.com/xpmir/cross-encoders)

cs.IR↗

arXiv2026

Answer Bubbles: Information Exposure in AI-Mediated Search

Generative search systems are increasingly replacing link-based retrieval with AI-generated summaries, yet little is known about how these systems differ in sources, language, and fidelity to cited material. We examine responses to 11,000 real search queries across five systems---vanilla GPT, Search GPT, Perplexity Search with Grok, Google AI Overviews, and traditional Google Search---at three levels: source diversity, linguistic characterization of the generated summary, and source-summary fidelity. We find that generative search systems exhibit significant \textit{source-selection} biases in their citations, favoring certain sources over others. Incorporating search also selectively attenuates epistemic markers, reducing hedging by up to 60\% while preserving confidence language in the AI-generated summaries. At the same time, AI summaries further compound the citation biases: Wikipedia and longer sources are disproportionately overrepresented, whereas cited social media content and negatively framed sources are substantially underrepresented. Our findings highlight the potential for \textit{answer bubbles}, in which identical queries yield structurally different information realities across systems, with implications for user trust, source visibility, and the transparency of AI-mediated information access.

cs.IR↗

arXiv2026

Semantic Trimming and Auxiliary Multi-step Prediction for Generative Recommendation

Generative Recommendation (GR) has recently transitioned from atomic item-indexing to Semantic ID (SID)-based frameworks to capture intrinsic item relationships and enhance generalization. However, the adoption of high-granularity SIDs leads to two critical challenges: prohibitive training overhead due to sequence expansion and unstable performance reliability characterized by non-monotonic accuracy fluctuations. We identify that these disparate issues are fundamentally rooted in the Semantic Dilution Effect, where redundant tokens waste massive computation and dilute the already sparse learning signals in recommendation. To counteract this, we propose STAMP (Semantic Trimming and Auxiliary Multi-step Prediction), a framework utilizing a dual-end optimization strategy. We argue that effective SID learning requires simultaneously addressing low input information density and sparse output supervision. On the input side, Semantic Adaptive Pruning (SAP) dynamically filters redundancy during the forward pass, converting noise-laden sequences into compact, information-rich representations. On the output side, Multi-step Auxiliary Prediction (MAP) employs a multi-token objective to densify feedback, strengthening long-range dependency capture and ensuring robust learning signals despite compressed inputs. Unifying input purification and signal amplification, STAMP enhances both training efficiency and representation capability. Experiments on public Amazon and large-scale industrial datasets show STAMP achieves 1.23--1.38$\times$ speedup and 17.2\%--54.7\% VRAM reduction while maintaining or improving performance across multiple architectures.

cs.IR↗

arXiv2026

A Wolf in Sheep's Clothing: Targeted Routing Hijacking in Federated RAG

Federated Retrieval-Augmented Generation (FedRAG) is attractive for privacy-sensitive applications because full local corpora remain on clients. As a result, routing must rely on client-provided semantic profiles, creating a new opportunity for manipulation. We introduce Routing Hijacking, a routing-stage attack in which a malicious client forges its profile to attract target queries despite having irrelevant underlying data. We show that this vulnerability is severe. Across three representative FedRAG routing architectures, Routing Hijacking consistently misroutes target queries and leads to downstream disruptions and failures, including missing evidence, poisoning, incorrect answers, and hallucinations. In a controlled MedQA-USMLE stress test, we further show that poisoned retrieved evidence can mislead models across scales, leading to incorrect answers, hallucinations, and sycophantic failures. Existing defenses do not close this gap: encrypted routing preserves the exploited ranking, and Byzantine-robust Federated Learning (FL) rules transfer poorly to heterogeneous routing profiles. To address this gap, we propose a trust-aware post-routing framework that reweights clients using returned-evidence feedback, including retrieval relevance, profile consistency, and cross-client agreement; online experiments show that it suppresses persistent hijacking over recurring queries and transfers to a learned neural router. Our findings establish routing integrity as a security challenge in FedRAG and highlight the need for stronger defenses for secure federated retrieval.

cs.CR↗

arXiv2026

ToolSense: A Diagnostic Framework for Auditing Parametric Tool Knowledge in LLMs

Large language models deployed as agents over large tool catalogs face a critical tool-retrieval bottleneck. As embedding-based retrieval approaches rely on compact encoders that may under-capture specialized tool semantics, parametric tool retrieval addresses this by encoding each tool as a virtual token appended to the LLM vocabulary, fine-tuned in two stages (memorization then retrieval SFT) to use the LLM as a retriever, achieving strong performance on standard ToolBench retrieval benchmarks. Yet these benchmarks use verbose, fully-specified queries, and their evaluation applies constrained decoding that restricts outputs to valid token paths, neither reveals whether the model actually understands its tools. We introduce \textbf{ToolSense}, an open-source LLM-powered diagnostic framework that takes any tool catalog as input and automatically generates three benchmarks: a Realistic Retrieval Benchmark (RRB) with queries at three ambiguity tiers, an MCQ probing benchmark, and a QA probing benchmark. Applying ToolSense to ToolBench (~47k tools) and evaluating five parametric model training configurations reveals a knowledge-retrieval dissociation: on RRB queries, several configurations collapse by ~50-64 percentage points compared to fully-specified ToolBench benchmarks, falling below the embedding-model baseline. Additionally, despite strong retrieval performance, some models score near-random on factual probes, suggesting a knowledge-retrieval dissociation. We open-source the ToolSense framework and the ToolBench diagnostic benchmarks at https://github.com/SAP/toolsense.

cs.AI↗

arXiv2026

Beyond Semantic IDs: Encoding Business-Value Ranking into Document Identifiers for Generative Retrieval

Generative Retrieval (GR) formulates retrieval as a sequence-to-sequence generation task, assigning each document a document identifier (DocID) and retrieving it through autoregressive decoding, making DocID design a critical factor in retrieval quality. However, existing schemes based on discrete representation learning suffer from inherent collision issues and create a mismatch between the DocID's encoding objective and the system's business optimization target. To address these limitations, we propose \textbf{Cluster-Ranked Identifier (CRID)}, which decouples DocID into \textit{semantic clustering} and \textit{business-value ranking}, yielding collision-free identifiers that support incremental updates via intra-cluster reranking. We further introduce an analytical framework that decomposes retrieval gains into \textit{personalized preference} and \textit{statistical prior} generalization, revealing how semantic cluster size governs the balance between the two components. Experiments on a Taobao e-commerce corpus of over 300M items show that CRID surpasses the strongest embedding-based retrieval baseline on top-K Hitrate, and delivers +1.06\% GMV in full-traffic deployment.

cs.IR↗

arXiv2026

REPREC: Representation Driven Parameter-Efficient Recommendation System

Large language models (LLMs) have been applied to sequential recommendation by incorporating collaborative signals through input conditioning or model adaptation. However, existing approaches often require LLM fine-tuning, additional architectural modules, representation distillation, or item-level conditioning over long interaction histories, increasing computational and deployment costs. We propose REPREC, a lightweight framework that conditions a frozen LLM using compact user-level representations. REPREC maps a fixed-size embedding from a frozen sequential encoder into a small set of learned soft tokens through an MLP injector, training only the injector while leaving both pretrained backbones unchanged. Our extensive experiments demonstrate that REPREC consistently improves recommendation performance across different sequential encoders, LLM backbones, and user activity levels. Its compact conditioning mechanism also makes REPREC computationally efficient during both training and inference. Moreover, training with short histories while evaluating with longer contexts retains 94--99\% of full-history performance while achieving an average $1.50\times$ per-epoch training speedup. The code is available at: https://github.com/phdbotcode/REPREC

cs.IR↗

arXiv2026

Search, Inspect, Fetch: Exploiting Structure-Aware Boolean Retrieval for Deep-Search Agents

Existing deep-search agents use a Search-Visit workflow that retrieves whole webpages without considering the structure they expose through titles, headings, sections, and metadata. This prevents agents from directly constraining retrieval to parts of a webpage and often carries irrelevant content into their context. We introduce Sieve, a search-inspect-fetch strategy driven by a Boolean Query Language (BQL): it searches webpage fields to filter candidates, uses an interchangeable ranker to order them, presents structure-rich result cards for inspection, and fetches only selected sections. Across three QA collections, Sieve is more accurate than the strongest conventional Search-Visit configuration on each collection while using 20.7-50.6% fewer tokens. Boolean filtering improves every tested ranker, and the accuracy-context advantage persists across retriever choices and agent backbones. Our implementation is included in the SkimSearchAgent library https://github.com/ielab/skim-search-agent.

cs.IR↗

arXiv2026

DocPC: Document-Level Visual Retrieval via Representative Page Composition

Visual document retrieval has advanced by encoding page screenshots with vision-language models, bypassing OCR pipelines. However, existing methods remain page-centric, misaligned with real-world scenarios requiring complete document retrieval. A naive page-then-document aggregation suffers from linear indexing cost and degraded retrieval when relevance spans multiple pages. We propose DocPC, a document-level visual retrieval framework based on Representative Page Composition: selecting representative pages and composing them into a single grid image for document-level indexing, reducing indexed images, vectors, and storage by 10.1x and end-to-end indexing time by roughly 7.7x. To handle multi-positive supervision prevalent at the document level, we combine multi-positive contrastive learning with sparsely scheduled listwise optimization. We also introduce DocViRe, a benchmark with multi-positive relevance annotations. DocPC-ColQwen achieves NDCG@5 of 44.09 on DocViRe, outperforming the strongest page-level baseline at 38.91 while reducing storage by 10.1x. Code is available at https://anonymous.4open.science/r/DocPC-Document-Level-Visual-Retrieval-via-Representative-Page-Composition-1D52. Data is available at https://huggingface.co/datasets/anonymous-7219/docpc.

cs.IR↗

arXiv2026

When Stale Constraints Go Unchecked: Budgeted Verification Failures in Inherited Agent Memory

Provenance links keep the evidence behind an inherited belief reachable; an agent with a verification budget must still choose which links to inspect. We study a consolidated memory that states a decision constraint and whose source record has since been superseded by a record that withdraws it: provenance is immutable, the current record has changed, and the memory is stale. In a controlled six-memory scenario with a budget of two records, sixteen language models rarely re-verified a constraint that read as settled: they inspected its provenance path in about one episode in five and, once the constraint had been superseded, produced stale-consistent decisions in 77.3%, 74.7% and 74.7% of episodes across a primary run, a replication and a held-out domain. Re-assigning one of the same two slots to the critical path removed most of them: +74.0, +72.7 and +61.3 points (positive in every model), +80.7 in a prospectively frozen interleaved replication with a repaired non-critical control, and +62.0 on a panel of 10 models from 9 organisations; a corrected re-run of the held-out scenario gave +73.3. The forced-critical policy uses experimenter knowledge of the critical path: it quantifies how much stale-decision risk the same budget can recover and is not a scheduler. Two further deposited experiments locate the failure and a remedy: in this store the constraint's path is selected in 17.0% of episodes at two slots and 88.7% at four of six (above uniform allocation), and at two slots a one-sentence, target-blind rule (prefer memories that state a limit on a candidate direction) moved the agent's own allocation onto the constraint's path and recovered the oracle contrast on decisions (+89.3 points) where that constraint limits the tempting action, while a content-free freshness cue did not materially redirect allocation and a content-matched control rule changed neither selection nor decisions.

cs.IR↗