Papers for
enterprise search developers
Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.
AutoOntic builds organized knowledge graphs automatically from text
AutoOntic: Automatic Ontology Induction from Text for Knowledge Graph Construction
Abstract: Graph-based retrieval-augmented generation depends on the quality of the knowledge graph (KG) it retrieves over, yet schema-free LLM-based extraction pipelines typically produce flat graphs: deduplicated entities and relations with no type hierarchy and no domain/range constraints. We present AutoOntic, a pipeline that automatically constructs an ontology from text by canonicalizing open-vocabulary entity types and relations, inducing a hierarchy over the canonicalized types, and mining relation-level domain and range constraints. Its hierarchy induction procedure prioritizes types using corpus-derived relation-argument profiles and places them through local, LLM-adjudicated parent/child checks, allowing newly inserted types to trigger reassignment of existing parent-child links. The resulting hierarchy constrains entity deduplication and supports the induction of relation-level domain/range constraints. For intrinsic evaluation, we introduce a reference-based protocol that classifies induced ontology types as exact, broader, narrower, or unrelated to reference domain and range classes. On the Wikidata-TekGen and DBpedia-WebNLG subsets of Text2KGBench, over 90% of AutoOntic's induced endpoint types are semantically aligned with the reference classes across two LLM backbones: about half match exactly, while most remaining differences concern granularity. On MINE, knowledge graphs organized using an ontology built by AutoOntic achieve 88% retrieval accuracy. AutoOntic is available at https://github.com/screemix/AutoOntic.
Controlled agents improve taxonomy building with adaptive tools and policies
Controlled Knowledge Engineering for Taxonomy Induction with Tool-Grounded LLM Agents and Runtime-Adaptive Policies
Abstract: Taxonomies are structured knowledge representations that support semantic search, recommendation, product catalog organization, and retrieval-augmented generation. Their automated construction from a fixed set of candidate entities requires more than plausible text generation: the resulting hierarchy must be complete, structurally valid, and semantically coherent. Large Language Models (LLMs) are promising for this task, but iterative free-form generation remains vulnerable to omitted or hallucinated entities, invalid hierarchical relations, and inconsistent organizational principles. We introduce TIGA (Taxonomy Induction with Tool-Grounded Agents), a knowledge-engineering architecture that formulates taxonomy induction as an iterative decision process over an explicit state and constrained action space. TIGA combines executable taxonomy-editing tools, an memory that records tool calls, rationales, and outcomes, and a runtime-adaptive organizational policy that the agent revises during inference. An optional global reviewer evaluates the constructed taxonomy against the candidate set and the active policy. We evaluate TIGA on four benchmark taxonomies using two LLM backbones and compare it with single-shot and iterative prompting-based baselines. TIGA achieves the strongest structural-quality results in most evaluated configurations, but does not dominate on every precision measure. The gains are accompanied by substantially higher token consumption, API-call volume, and inference time. Ablation results show consistent benefits from the policy and interaction-memory components, whereas the effect of global review depends on the dataset and the available revision budget. These findings characterize tool grounding and runtime policy adaptation as promising but costly mechanisms for improving the controlled construction of structured knowledge representations.
Instruction following retrieval improves with natural hard negatives
Natural Hard-Negative Mining from Positives for Instruction-Following Retrieval
Abstract: Instruction-following information retrieval (IFIR) requires retrievers to rank documents by both a query and its instance-specific instructions, yet instruction-conditioned contrastive supervision remains scarce. Existing synthetic methods typically fix a positive passage and generate additional negatives, overlooking the hard-negative potential of the original positive. We propose Natural Hard-Negative Mining (NHNM), which promotes convertible negatives under query-compatible instructions and repurposes original positives as natural instruction hard negatives. Its automated Candidate Cleaning, Reversal Construction, and Geometry Filtering pipeline produces compact, high-contrast supervision aligned with the pretrained retrieval space. Applied to MSMARCO, NHNM yields NHNM-MARCO, an IFIR training set of nearly 7K instances. On InstructIR and FollowIR, models with 568M to 4B parameters fine-tuned on NHNM-MARCO achieve strong instruction-following performance and match or surpass substantially larger retrievers in several settings. Controlled analyses and ablations attribute these gains to natural hard-negative supervision rather than model scale or training-data volume.
Programmable search improves complex question answering and reasoning
Learning to Program Retrievals: Composing Search Primitives via Reinforcement Learning
Abstract: Retrieval-augmented language models increasingly use multi-step search to solve knowledge-intensive tasks, yet existing systems typically expose retrieval through fixed pipelines or collections of atomic tools. We introduce Retrieval Program Learning (RPL), a programmable search framework that allows a policy to compose search primitives into executable programs with local state and control flow. This enables adaptive search that can retrieve broadly, process intermediate results, and selectively construct model context without tying retrieval breadth to context consumption. However, the resulting flexibility creates a difficult exploration problem: useful programs may require several coordinated decisions, while reinforcement learning often supplies only sparse feedback. To address this challenge, we propose program mutation, a grammar-constrained and execution-verified procedure that generates local variants of suboptimal programs sampled during RL, converts them into training trajectories, and distills them back into the policy. We evaluate RPL on open-domain question answering and corpus-level reasoning tasks, comparing it with a range of retrieval and reasoning baselines. RPL consistently improves open-domain QA performance across model sizes and in- and out-of-domain benchmarks, with particularly large gains on multi-hop tasks. Notably, a 4B model trained with RPL outperforms all 8B model baselines. On corpus-level reasoning, RPL achieves the highest answer F1 and document-coverage F1 for both model sizes while substantially reducing context consumption and retrieval turns. Our results show that programmable search expands the retrieval strategies that agents can learn, while program mutation provides an effective mechanism for exploring useful program structures beyond direct on-policy sampling.
Memory retrieval improves by using neighborhood-aware residuals
NEAR: Neighborhood-Aware Residuals for Memory Retrieval
Abstract: Memory retrieval aims to identify evidence from stored records to answer a query. Stored records with different factual details may share topics, entities, or events. Such shared contents would make records without the required evidence more likely to receive high similarity scores and outrank the correct evidence. In this paper, we use the information shared by these records to construct local references for query-memory matching. We propose NEighborhood-Aware Residuals (NEAR), a training-free framework that uses these references to construct residual representations that complement the original embeddings. Specifically, each memory derives a reference direction from its neighbors, and each query obtains its reference from a precomputed table. Using these references, NEAR constructs projection residuals for both queries and memories. These residuals are combined with the original embeddings for a single approximate nearest-neighbor search over a fixed memory index. Experiments on eight long-context retrieval benchmarks show consistent improvements across nine embedding models. Downstream evaluations show that NEAR improves question-answering performance.
Generated query expansion improves strong sparse search systems performance
Generated Query Expansion Still Helps Strong Sparse Retrieval: A Controlled Study with SPLADE-v3
Abstract: Scientific queries are often brief, while relevant papers use specialized vocabulary. Generated query expansion can bridge this mismatch, but earlier work suggests that its value shrinks as the underlying retriever becomes stronger. We test the four generated formats of term lists, a pseudo-document, multiple pseudo-references, and corpus-steered text all together with SPLADE-v3 on NFCorpus, TREC-COVID, and SciDocs. Every condition searches the same frozen document index and follows the same query-side integration rule and 256-dimension budget, isolating the effect of the added content. All twelve method-collection comparisons improve aggregate nDCG@10, with best relative gains of 4.81%, 8.92%, and 9.47%. Eleven remain significant after Holm correction. The gain persists in 103 of 114 interpolation settings, including every setting that assigns at least 30% of the mixture weight to the original query. Shuffled-text and non-contextual lexical-bag controls also remain above baseline in all 24 aggregate comparisons, showing that the added vocabulary carries most of the benefit. A corpus-induced typed concept graph, by contrast, produces no consistent gain, and its relation, depth, validation, random, and gating controls do not rescue it. Generated vocabulary can therefore complement a strong learned sparse retriever, provided that the original query remains strongly represented.
Efficient dense vector search boosts multimodal querying in knowledge graphs
Efficient Dense Vector Search within Knowledge Graph Content Embeddings
Abstract: Knowledge graphs are a core component of today's knowledge infrastructure, supporting reasoning and anchoring knowledge systems to verifiable facts. RDF stores and SPARQL engines fulfill this function, enabling a range of retrieval and inference tasks on structured knowledge. Coupling them with Language Models (LMs) extends RAG toward neurosymbolic reasoning, where structured queries gate or re-rank generative outputs. This line of reasoning requires that SPARQL evaluation natively support tensor operations on dense embeddings, enabling multimodal querying and learned similarity-based ranking to be expressed together with graph-structural constraints. This approach is feasible only if the engine can efficiently perform dense vector search. We present QLever-Unified Indexed Vector Embedding Retrieval (QUIVER), an extension to QLever that adds native support for dense vector retrieval within RDF knowledge graphs. It implements three optimizations: engine-level registration of tensor functions, vocabulary-time parsing of JSON-encoded vectors, and a virtual SERVICE that exposes a vector index inside the query. We propose two new benchmarks: an extension of Berlin Sparql Benchmark (BSBM) with text embeddings and an extension of DBpedia with image embeddings. Against the baselines, vocabulary-time parsing alone yields median speedups of up to 41.9x on BSBM and 20x on DBpedia for single-type ranking; adding an approximate nearest-neighbor index yields speedups of 355x on BSBM and 97.8x on DBpedia. The index further makes cross-modal vector joins on DBpedia feasible in seconds, whereas all non-indexed configurations time out.
Mixture-of-experts language models improve search speed and accuracy
Mixture-of-Experts Language Models Can Be Strong and Efficient Retrievers
Abstract: Recent work has shown that fine-tuning decoder-only large language models (LLMs) for retrieval yields strong first-stage retrievers, with effectiveness improving as backbones grow in size. However, every query and document must pass through the full model, so encoding cost increases with model size. Mixture-of-Experts (MoE) LLMs activate only a subset of parameters per token and are widely used to scale generative models, yet remain underexplored as retrievers. We systematically study MoE backbones for retrieval by training MoE and dense LLMs from several families using the same procedure, evaluating them across diverse datasets, and measuring query encoding time under the same serving configuration. We show that MoE retrievers outperform dense retrievers with comparable active parameter counts by up to 3.0 nDCG@10 points on BEIR. One of our strongest MoE retrievers matches an 8B dense retriever with 59% fewer active parameters and 18% lower query encoding time. We further show that the number of experts used for query encoding can be reduced without retraining or re-indexing, retaining more than 99% of retrieval effectiveness while reducing query encoding time by up to 26%. Recent rerankers provide only modest additional gains over strong MoE first stages, which often match or exceed the reranked configurations we evaluate. Together, these results show that MoE LLMs can be strong and efficient first-stage retrievers.
Visual document retrieval improves by picking tokens after query arrives
Query-Aware Token Budgeting for Efficient Late-Interaction Visual Document Retrieval
Abstract: Late-interaction visual document retrievers preserve fine-grained page evidence by storing many token embeddings per page, but the resulting storage and query-time interaction costs make large-scale deployment expensive. Pooling document tokens before indexing offers a natural remedy, yet static pooling must decide which visual evidence to preserve before the query is known. We study an alternative: a heavily compressed hot-path index generates candidates, after which query-aware token budgeting operates on the original token sets of the shortlisted pages. We formulate this stage-two selection as a budgeted MaxSim coverage problem, show that a clipped version is monotone submodular, and compare coverage-only, cluster-guided, token-wise, and marginal-gain policies. On ten ViDoRe tasks with ColModernVBERT, direct static pooling reduces macro normalized discounted cumulative gain at rank five from 0.6309 without compression to 0.4738 at a thirty-two-fold pool factor. Under the same candidate-generation regime and a pool-factor-eight-equivalent reranking budget, token top-k recovers 93.93 percent of the full-token score, while greedy marginal-gain selection recovers 98.39 percent. Held-out and leave-one-dataset-out evaluations yield positive greedy improvements over token top-k on every dataset. The latency analysis reveals two useful operating points: token top-k for interactive retrieval and the naive greedy implementation as a quality upper envelope. Together, these results show that late-interaction visual retrieval benefits from query-aware allocation rather than query-agnostic pooling alone.