# Entity resolution

Entity resolution is a data-processing technology for matching and linking related records across applications, channels, and data stores.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Can LLMs Resolve Entity Conflicts in Domain Specific NLU Pipelines?

DevFeed: [Can LLMs Resolve Entity Conflicts in Domain Specific NLU Pipelines?](<https://devfeed.tech/articles/can-llms-resolve-entity-conflicts-in-domain-specific-nlu-pipelines-145095.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25733>)

Author: Akansha Bhardwaj, Mingbo Cui, Victor Guimaraes, Dominik Kohler, Diana Nicoleta Popa, Meghana M. Reddy, Paolo Rosso

Published: 2026-11-20T05:00:00Z

Content type: comparison

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>), [lora](<https://devfeed.tech/topics/lora.md>), [Enhancing AI model performance](<https://devfeed.tech/topics/enhancing-ai-model-performance.md>)

Tags: [cases](<https://devfeed.tech/tags/cases.md>), [comparison](<https://devfeed.tech/tags/comparison.md>), [complex](<https://devfeed.tech/tags/complex.md>), [conflict-resolution](<https://devfeed.tech/tags/conflict-resolution.md>), [cs-ir](<https://devfeed.tech/tags/cs-ir.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [fine-tuning](<https://devfeed.tech/tags/fine-tuning.md>), [flexibility](<https://devfeed.tech/tags/flexibility.md>), [llama3](<https://devfeed.tech/tags/llama3.md>), [llms](<https://devfeed.tech/tags/llms.md>), [lora](<https://devfeed.tech/tags/lora.md>), [pipelines](<https://devfeed.tech/tags/pipelines.md>), [prompting](<https://devfeed.tech/tags/prompting.md>), [strategies](<https://devfeed.tech/tags/strategies.md>)

### AI overview

The study compares rule-based entity conflict resolution with two Llama3.1-8B-Instruct approaches in domain-specific NLU pipelines: multi-shot prompting and LoRA fine-tuning. It evaluates how well the models match existing pipeline outputs and, separately, their accuracy on human-labeled data. The abstract reports that prompting is easy to adapt and allows rapid behavioral changes, while fine-tuning performs better on complex, ambiguous cases. It suggests combining the approaches to pair prompting's flexibility with fine-tuning's broader accuracy gains.

### Source excerpt

arXiv:2611.25733v1 Announce Type: cross Abstract: Resolving entity conflicts is a critical step in Pipeline-based NLU architectures, especially when multiple extractors provide overlapping or conflicting outputs. Traditional rule-based entity conflict resolution methods often lack flexibility and are unscalable. In this work, we aim to study if the broad knowledge of LLMs can be leveraged to resolve entity conflicts in domain specific NLU pipelines. We present an experimental comparison between a traditional rule-based conflict resolution method, and two LLM-based approaches: a multi-shot prompting method using the Llama3.1-8B-Instruct model and a LoRA fine-tuned Llama3.1-8B-Instruct model. The evaluation was conducted under two setups: (1) a broad assessment of LLM's capability to match existing NLU pipeline output, and (2) a more controlled scenario assessing the LLM accuracy on human-labeled data. Results show that while multi-shot prompting offers ease of adaptation and rapid behavioral changes, the fine-tuned model delivers superior performance on complex, ambiguous cases. Ultimately, the best approach appears to be a combination of both strategies, leveraging multi-shot prompting's flexibility alongside the broad accuracy gains of LoRA fine-tuning for a more comprehensive performance.

## MECI: Multi-Element Collaborative Interaction for Multimodal Entity Linking

DevFeed: [MECI: Multi-Element Collaborative Interaction for Multimodal Entity Linking](<https://devfeed.tech/articles/meci-multi-element-collaborative-interaction-for-multimodal-entity-linking-142806.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24522>)

Author: Jie Peng, Yongxue Shan, Yongfu Zha, Xiaodong Wang

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [Mixture of Experts (MoE)](<https://devfeed.tech/topics/mixture-of-experts-moe.md>)

Tags: [collaboration](<https://devfeed.tech/tags/collaboration.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [cross](<https://devfeed.tech/tags/cross.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [divide-and-conquer](<https://devfeed.tech/tags/divide-and-conquer.md>), [element](<https://devfeed.tech/tags/element.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [experts](<https://devfeed.tech/tags/experts.md>), [feature](<https://devfeed.tech/tags/feature.md>), [framework](<https://devfeed.tech/tags/framework.md>), [generative](<https://devfeed.tech/tags/generative.md>), [global](<https://devfeed.tech/tags/global.md>), [grounding](<https://devfeed.tech/tags/grounding.md>), [interaction](<https://devfeed.tech/tags/interaction.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [knowledge-base](<https://devfeed.tech/tags/knowledge-base.md>), [language](<https://devfeed.tech/tags/language.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [large](<https://devfeed.tech/tags/large.md>), [linking](<https://devfeed.tech/tags/linking.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>)

### AI overview

MECI addresses element dominance in multimodal entity linking with decoupled feature learning and hierarchical cross-modal interaction. Its Multi-view Experts Network uses KL-guided routing to coordinate specialized experts, while a modality-aware weighting network combines evidence across semantic levels. A final generative refinement stage uses large language models for zero-shot re-ranking. On WikiDiverse, RichpediaMEL, and WikiMEL, the abstract reports Hits@1 improvements of 1.95%, 7.30%, and 2.31%, respectively, over state-of-the-art baselines.

### Source excerpt

arXiv:2611.24522v1 Announce Type: cross Abstract: Multimodal Entity Linking (MEL) aims to disambiguate mentions in multimodal contexts by grounding them to specific entities in a knowledge base. A pivotal challenge in MEL is capturing multi-level correspondences: the semantic consistency between mention-entity pairs and the complementary correlations across modalities. However, existing methods often suffer from element dominance due to their reliance on coupled interactions or coarse global aggregations. In response, we propose the Multi-Element Collaborative Interaction (MECI) framework. First, to capture multi-element mention-entity correspondences, we develop a Multi-view Experts Network that leverages a "divide-and-conquer" strategy for decoupled feature learning to mitigate element dominance, supported by a KL-guided routing mechanism that governs expert specialization and collaboration. Furthermore, to model cross-modal complementary correlations, we propose a Hierarchical Multimodal Interaction Module, where a dynamic modality-aware weighting network refines interactions across hierarchical semantic levels, thereby integrating multi-granular evidence to counteract element dominance. Finally, we incorporate a generative semantic refinement stage that utilizes large language models for zero-shot re-ranking. Extensive experiments on WikiDiverse, RichpediaMEL, and WikiMEL show that MECI consistently outperforms state-of-the-art baselines, improving Hits@1 by 1.95%, 7.30%, and 2.31%, respectively.

## Growing Clusters Lower the Bar: Profile Accretion in LLM-Based Incremental Entity Resolution

DevFeed: [Growing Clusters Lower the Bar: Profile Accretion in LLM-Based Incremental Entity Resolution](<https://devfeed.tech/articles/growing-clusters-lower-the-bar-profile-accretion-in-llm-based-incremental-entity-resolution-140632.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22485>)

Author: Huixue Han, Shangzhi Xu, Anqi Xiong, Yu-Chu Tian

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Databases](<https://devfeed.tech/sources/arxiv-databases.md>)

Topics: [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [acceptance](<https://devfeed.tech/tags/acceptance.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [blocks](<https://devfeed.tech/tags/blocks.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-sonnet](<https://devfeed.tech/tags/claude-sonnet.md>), [claude-sonnet-5](<https://devfeed.tech/tags/claude-sonnet-5.md>), [cluster](<https://devfeed.tech/tags/cluster.md>), [clusters](<https://devfeed.tech/tags/clusters.md>), [company](<https://devfeed.tech/tags/company.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-db](<https://devfeed.tech/tags/cs-db.md>), [data](<https://devfeed.tech/tags/data.md>), [entity-resolution](<https://devfeed.tech/tags/entity-resolution.md>), [factor](<https://devfeed.tech/tags/factor.md>), [formats](<https://devfeed.tech/tags/formats.md>), [haiku](<https://devfeed.tech/tags/haiku.md>), [incremental](<https://devfeed.tech/tags/incremental.md>), [intruder](<https://devfeed.tech/tags/intruder.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [large-language-model](<https://devfeed.tech/tags/large-language-model.md>), [llm](<https://devfeed.tech/tags/llm.md>)

### AI overview

In incremental entity resolution, tested LLM matchers became more willing to accept a new record as cluster size grew. On French procurement suppliers, showing 16 verified members instead of one increased false matches among similarly named companies in six open-weight models, while AUC rose only slightly; the shift also appeared with addresses, prompt variants, and BatchER and ComEM formats. A false merge further increased acceptance of the intruder's other records in every open model. The effect was absent on Czech and Bulgarian TED notices, where a record with its full address already identified the company. For online linking on confusable blocks, one threshold calibrated on labelled data increased recall for five models without detectable precision loss or additional LLM calls. The abstract's reported numeric ranges are unresolved placeholders.

### Source excerpt

arXiv:2611.22485v1 Announce Type: cross Abstract: Incremental entity resolution pipelines ask a large language model (LLM) whether a new record belongs to an existing cluster and show it the records already linked. Analyses of matching oracles assume that the answer depends only on the records compared. For the LLM matchers we test, it depends on cluster size, and the model lowers its bar for joining a cluster as the cluster grows. On French procurement suppliers from FOPPA, with SIREN identifiers as ground truth, we hold the new record fixed and vary only the number of verified members shown. Sixteen members instead of one raise the false match rate on similarly named companies by \dfmrMin--\dfmrMax{} percentage points in six open-weight models, while their area under the ROC curve rises by only \aucGainMin--\aucGainMax. In Claude Sonnet~5 and Haiku~5.5 the rise is \clSonD{} and \clHaiD{} points. The shift persists with street addresses, prompt variants and the BatchER and ComEM formats, and one false merge raises acceptance of the intruder's other records in every open model. It does not arise on Czech and Bulgarian TED notices, where one record with its full address already identifies the company. In online linking on confusable blocks, a single threshold set on labelled data raises recall by a factor of \fxRRmin--\fxRRmax{} for five models with no detectable loss of precision and no extra LLM calls.

## Fuzzy string matching is now generally available across Microsoft SQL

DevFeed: [Fuzzy string matching is now generally available across Microsoft SQL](<https://devfeed.tech/articles/fuzzy-string-matching-is-now-generally-available-across-microsoft-sql-140689.md>)

Original publisher: [Read original article](<https://devblogs.microsoft.com/azure-sql/fuzzy-string-matching-is-now-generally-available-across-microsoft-sql/>)

Author: Abhiman Tiwari

Published: 2026-11-17T20:54:16Z

Content type: article

Language: en

Sources: [Azure SQL Dev Corner](<https://devfeed.tech/sources/azure-sql-dev-corner.md>)

Topics: [sql-server](<https://devfeed.tech/topics/sql-server.md>), [SQL](<https://devfeed.tech/topics/sql.md>), [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [Azure SQL](<https://devfeed.tech/topics/azure-sql.md>)

Tags: [azure-sql](<https://devfeed.tech/tags/azure-sql.md>), [azure-sql-database](<https://devfeed.tech/tags/azure-sql-database.md>), [azure-sql-managed-instance](<https://devfeed.tech/tags/azure-sql-managed-instance.md>), [edit-distance](<https://devfeed.tech/tags/edit-distance.md>), [fabric-sql-database](<https://devfeed.tech/tags/fabric-sql-database.md>), [fuzzy-string-matching](<https://devfeed.tech/tags/fuzzy-string-matching.md>), [managed-instance](<https://devfeed.tech/tags/managed-instance.md>), [microsoft](<https://devfeed.tech/tags/microsoft.md>), [microsoft-sql](<https://devfeed.tech/tags/microsoft-sql.md>), [native](<https://devfeed.tech/tags/native.md>), [similarity](<https://devfeed.tech/tags/similarity.md>), [sql-database-in-fabric](<https://devfeed.tech/tags/sql-database-in-fabric.md>), [sql-server-2025](<https://devfeed.tech/tags/sql-server-2025.md>), [string-matching](<https://devfeed.tech/tags/string-matching.md>), [t-sql](<https://devfeed.tech/tags/t-sql.md>)

### AI overview

Microsoft SQL now offers four native T-SQL fuzzy matching functions across Azure SQL Database, Azure SQL Managed Instance, SQL database in Microsoft Fabric, and SQL Server 2025 CU11 or later. `EDIT_DISTANCE` uses restricted Damerau-Levenshtein (Optimal String Alignment) to count insertions, deletions, substitutions, and transpositions; its optional distance limit returns the actual count within the limit and limit + 1 when exceeded. `EDIT_DISTANCE_SIMILARITY` normalizes that count. Jaro-Winkler functions score character resemblance, including a fixed shared-prefix preference, so they can rank names but may score distinct organizations highly. These scores indicate resemblance, not identity or probability. The article recommends narrowing candidates with trusted attributes before comparing, validating cutoffs on labeled data, and confirming likely matches with independent evidence. Inputs require supported Windows or binary collations, bounded character types, and return NULL for NULL inputs; `varchar(max)` and `nvarchar(max)` are unsupported.

### Source excerpt

Four native T-SQL functions that tell you how close two strings are, not just whether they're equal. MICROSFT isn't MICROSOFT. MARHTA isn't MARTHA. INV-1O08 isn't INV-1008. To an equality check, each pair is simply different. To your business, each is a near miss worth catching. Today, fuzzy string matching is generally available in Microsoft SQL. EDIT_DISTANCE tells you MICROSFT is one edit from MICROSOFT. JARO_WINKLER_SIMILARITY scores MARTHA and MARHTA at 96 out of 100. [...] The post Fuzzy string matching is now generally available across Microsoft SQL appeared first on Azure SQL Dev Corner.

## MAPS: A Multimodal Agentic Framework for POI Data Production Based on Crowdsourced Street-view Imagery

DevFeed: [MAPS: A Multimodal Agentic Framework for POI Data Production Based on Crowdsourced Street-view Imagery](<https://devfeed.tech/articles/maps-a-multimodal-agentic-framework-for-poi-data-production-based-on-crowdsourced-street-view-imagery-137496.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19807>)

Author: Sijie Ruan, Tingyu Xie, Mingfei Liu, Yicheng Tang, Haoyu Han, Jingnan Qi, Jingshu Zeng, Dongqi Xu, Jing Wang, Shuliang Wang

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Databases](<https://devfeed.tech/sources/arxiv-databases.md>)

Topics: [smolvlm](<https://devfeed.tech/topics/smolvlm.md>), [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [geospatial](<https://devfeed.tech/topics/geospatial.md>), [Image](<https://devfeed.tech/topics/image.md>), [managing large-scale text data in AI](<https://devfeed.tech/topics/managing-large-scale-text-data-in-ai.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [actions](<https://devfeed.tech/tags/actions.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [also](<https://devfeed.tech/tags/also.md>), [annotation](<https://devfeed.tech/tags/annotation.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [based](<https://devfeed.tech/tags/based.md>), [but](<https://devfeed.tech/tags/but.md>), [candidate](<https://devfeed.tech/tags/candidate.md>), [changes](<https://devfeed.tech/tags/changes.md>), [completeness](<https://devfeed.tech/tags/completeness.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-db](<https://devfeed.tech/tags/cs-db.md>), [data](<https://devfeed.tech/tags/data.md>), [database](<https://devfeed.tech/tags/database.md>), [databases](<https://devfeed.tech/tags/databases.md>), [framework](<https://devfeed.tech/tags/framework.md>), [imagery](<https://devfeed.tech/tags/imagery.md>), [maps](<https://devfeed.tech/tags/maps.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [production](<https://devfeed.tech/tags/production.md>), [view](<https://devfeed.tech/tags/view.md>)

### AI overview

MAPS turns crowdsourced street-view imagery into POI database updates through two cascaded modules. SPICE adapts a lightweight vision-language model to reconstruct and filter text, group entries by POI, and classify their meaning and completeness. RIMA retrieves candidate records, matches POIs using evidence and reasons about attributes, selectively gathering additional visual, geospatial, or external-source evidence when needed. In a human-reviewed full-task evaluation, MAPS outperformed the strongest baseline by 21.8%; deployment in Kuaishou's large-scale POI pipeline reduced average processing time by 76.9% compared with manual annotation.

### Source excerpt

arXiv:2611.19807v1 Announce Type: new Abstract: Point-of-interest (POI) databases support location-based services, yet urban changes quickly make their records stale and manual maintenance is costly. Crowdsourced street-view images provide timely visual evidence, but converting them into database actions requires separating multiple POIs and irrelevant text in each image, then resolving incomplete observations against ambiguous records. We study Street-view Image-based POI Data Production and propose MAPS, the first framework to integrate structured POI extraction with database maintenance from such imagery. MAPS contains two cascaded modules. SPICE progressively adapts a lightweight vision-language model to reconstruct and filter text, group entries belonging to the same POI, and classify their semantics and completeness. RIMA retrieves candidate records, performs evidence-grounded POI matching and attribute-level reasoning, and selectively acquires complementary visual, geospatial, or external-source evidence when necessary. In human-reviewed full-task evaluation, MAPS outperforms the strongest baseline by 21.8\%. MAPS has also been deployed in Kuaishou's large-scale POI production pipeline, reducing average processing time by 76.9\% relative to manual annotation. These results demonstrate the effectiveness and practical value of MAPS for scalable POI maintenance.

## StaleBench: Isolating Stale Retrieval in Agent Memory, and Pricing the Fix

DevFeed: [StaleBench: Isolating Stale Retrieval in Agent Memory, and Pricing the Fix](<https://devfeed.tech/articles/stalebench-isolating-stale-retrieval-in-agent-memory-and-pricing-the-fix-137491.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19752>)

Author: S M Mahbub Murshed

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [search-engines](<https://devfeed.tech/topics/search-engines.md>), [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-memory](<https://devfeed.tech/tags/agent-memory.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-ir](<https://devfeed.tech/tags/cs-ir.md>), [memory](<https://devfeed.tech/tags/memory.md>), [retrieval](<https://devfeed.tech/tags/retrieval.md>)

### AI overview

StaleBench evaluates stale retrieval across 892 real updating facts, using validity labels from Wikidata via TempLAMA rather than retrieval scores. It finds that retrieval metrics can misrank memory policies: a retention policy and a retain-everything control returned identical rank-1 statements, yet reader accuracy differed by 29.5 and 65.6 points when dates were hidden. The new Contamination@5 metric ranks these policies correctly where current@1 and current@5 do not. Removing statements' lexical anchors reduced freshness for deployed keying mechanisms by tens of points; entity resolution during both writing and querying recovered most of that loss. Mem0, Graphiti, LangMem, and a reimplemented ERASE handle supersession differently, with no single scoring convention describing all four. The authors release the data, frozen evaluator, and reference baselines.

### Source excerpt

arXiv:2611.19752v1 Announce Type: cross Abstract: An agent that stores facts over time eventually stores several versions of the same fact. Asked a value-agnostic question - "which team does X play for now?" - a similarity retriever cannot prefer the current version, because every version matches the entity the query names. StaleBench isolates this stale-retrieval failure on 892 real updating facts, with currency labels from the source (Wikidata validity years via TempLAMA), never from a retrieval score. Our central result is that retrieval metrics can rank memory policies backwards: a retention policy and its retain-everything control return the identical statement at rank 1 for every test query, yet differ by 29.5 and 65.6 points in reader accuracy on two readers when no date is shown. Contamination@5, the metric we add, ranks them correctly where current@1 and current@5 do not. Withdrawing the lexical anchor that statements carry costs deployed keying mechanisms tens of points of freshness; entity resolution on both the write path and the query recovers most of the loss. Three deployed systems (Mem0, Graphiti, LangMem) and a reimplemented ERASE each handle supersession differently, and no single scoring convention describes all four. We release the data, a frozen evaluator, and reference baselines.

## Real-World End-to-End Product Entity Resolution

DevFeed: [Real-World End-to-End Product Entity Resolution](<https://devfeed.tech/articles/real-world-end-to-end-product-entity-resolution-138295.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19418>)

Author: Dimitrios Karapiperis, George Papadakis, Manolis Koubarakis

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Databases](<https://devfeed.tech/sources/arxiv-databases.md>)

Topics: [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>)

Tags: [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [blocking](<https://devfeed.tech/tags/blocking.md>), [brand](<https://devfeed.tech/tags/brand.md>), [cascade](<https://devfeed.tech/tags/cascade.md>), [cross-encoder](<https://devfeed.tech/tags/cross-encoder.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-db](<https://devfeed.tech/tags/cs-db.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [e-commerce](<https://devfeed.tech/tags/e-commerce.md>), [entity-resolution](<https://devfeed.tech/tags/entity-resolution.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [formatting](<https://devfeed.tech/tags/formatting.md>), [fusion](<https://devfeed.tech/tags/fusion.md>), [generative](<https://devfeed.tech/tags/generative.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [large-language-model](<https://devfeed.tech/tags/large-language-model.md>), [lexical](<https://devfeed.tech/tags/lexical.md>), [memory](<https://devfeed.tech/tags/memory.md>), [model](<https://devfeed.tech/tags/model.md>), [product](<https://devfeed.tech/tags/product.md>), [real-world](<https://devfeed.tech/tags/real-world.md>)

### AI overview

The study evaluates four zero-shot product entity resolution pipelines--Lexical Cascade, Semantic Cascade, Hybrid Fusion Cascade, and Generative-Lexical Cascade--on established benchmarks and two newly introduced real-world datasets with cross-lingual noise and dense brand collisions. It compares lexical and dense blocking paired with cross-encoder, tensor-search, or large language model verification, measuring accuracy, runtime, and memory consumption. The supplied excerpt gives the evaluation setup but no results showing which approaches perform best.

### Source excerpt

arXiv:2611.19418v1 Announce Type: cross Abstract: Product Entity Resolution (PER) in e-commerce is challenging due to unstructured text, formatting inconsistencies, promotional noise, and fine-grained product differences. We present an end-to-end evaluation of four zero-shot PER pipelines---Lexical Cascade, Semantic Cascade, Hybrid Fusion Cascade, and Generative-Lexical Cascade---across established benchmarks and introduce two real-world datasets characterized by cross-lingual noise and dense brand collisions. We evaluate lexical and dense blocking with cross-encoder, tensor-search, and Large Language Model verification, considering accuracy, runtime, and memory consumption.

## Exact Sparse Assignment with Costly Target Embeddings: Fixed-Realization Certificates and an Empirical Audit

DevFeed: [Exact Sparse Assignment with Costly Target Embeddings: Fixed-Realization Certificates and an Empirical Audit](<https://devfeed.tech/articles/exact-sparse-assignment-with-costly-target-embeddings-fixed-realization-certificates-and-an-empirical-audit-138928.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19770>)

Author: Hamidreza Hasani Balyani, Mahtab Movahhedrad, Seyed Pouyan Mousavi Davoudi

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [audit](<https://devfeed.tech/topics/audit.md>)

Tags: [audit](<https://devfeed.tech/tags/audit.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [certificates](<https://devfeed.tech/tags/certificates.md>), [cs-ir](<https://devfeed.tech/tags/cs-ir.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [matching](<https://devfeed.tech/tags/matching.md>), [neural](<https://devfeed.tech/tags/neural.md>), [runtime](<https://devfeed.tech/tags/runtime.md>)

### AI overview

The paper gives a certificate for stopping selective target-embedding acquisition once a sparse optional assignment is optimal for the acquired feature values and every unit-ball completion of the remaining targets. In tests on four text entity-matching benchmarks with two frozen encoders, all 96 certificate simulations and 120 online timings were exact for their own realization. At group size 8, the shared certificate acquired 60% of candidate-active targets on average, beating mean-centre and MEB bounds but matching the zero-centre bound; savings were concentrated in one benchmark. Selective execution was 0.510 seconds slower per run on average (95% interval 0.248-0.772 seconds slower). Re-encoding in uniform microbatches preserved the optimum in only 5 of 8 encoder-dataset cells, and 45 of 120 online runs differed from an eager cache at near-ties. The certificate concerns the realized acquired rows, not features from a differently batched forward pass.

### Source excerpt

arXiv:2611.19770v1 Announce Type: new Abstract: Embedding-based one-to-one association is common in vision and data integration: detections are associated with tracks and records are linked across databases. Such pipelines often solve a sparse optional assignment whose edge scores add a cheap term to an inner product of neural embeddings. When target embeddings are costly, one can acquire them selectively and stop once a certificate proves that the current assignment is optimal. We give such a certificate for unit-norm embeddings. It bounds the unknown part of every reassignment by a support function and maximizes the resulting robust gains with a second sparse assignment; a shared owner-centred bound dominates any fixed-centre triangle bound. We make precise what is certified when network features are not batch-invariant: the realized values of the acquired rows and every unit-ball completion of the rest, not a second forward pass in a different batch. Auditing against a separately batched cache tests a different property; on our first pilot it produced a spurious $2.49\times10^{-6}$ objective failure. We audit the corrected protocol on four public text entity-matching benchmarks with two frozen encoders; no image data is used. All 96 certificate simulations and 120 online selective timings were exact on their own realization (largest objective gap $6.82\times10^{-13}$). At group size 8 the shared certificate acquires on average 60% of candidate-active targets, against 74% (mean centre), 69% (MEB) and 60% (zero centre): it beats the mean and MEB centres but not the zero centre, and savings concentrate in one benchmark (7% acquired). The prespecified runtime claim is contradicted: selective execution is +0.510 s slower per run (dataset-cluster 95% interval [+0.248, +0.772]). Uniform-microbatch re-encodings preserved the optimum in only 5 of 8 encoder-dataset cells, and 45 of 120 online runs disagreed with an eager cache at near-ties.

## How Thermo Fisher Scientific built AI-powered patient matching on AWS

DevFeed: [How Thermo Fisher Scientific built AI-powered patient matching on AWS](<https://devfeed.tech/articles/how-thermo-fisher-scientific-built-ai-powered-patient-matching-on-aws-133534.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/architecture/how-thermo-fisher-scientific-built-ai-powered-patient-matching-on-aws/>)

Author: Kaushal Patel

Published: 2026-11-13T20:26:58Z

Content type: article

Language: en

Sources: [AWS Architecture Blog](<https://devfeed.tech/sources/aws-architecture-blog.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Amazon Web Services](<https://devfeed.tech/topics/aws.md>), [AWS Glue](<https://devfeed.tech/topics/aws-glue.md>), [AWS Step Functions](<https://devfeed.tech/topics/aws-step-functions.md>), [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [database-optimization](<https://devfeed.tech/topics/database-optimization.md>)

Tags: [aggregate](<https://devfeed.tech/tags/aggregate.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-powered](<https://devfeed.tech/tags/ai-powered.md>), [ai-workloads](<https://devfeed.tech/tags/ai-workloads.md>), [amazon-athena](<https://devfeed.tech/tags/amazon-athena.md>), [amazon-documentdb](<https://devfeed.tech/tags/amazon-documentdb.md>), [amazon-elastic-container-service](<https://devfeed.tech/tags/amazon-elastic-container-service.md>), [amazon-s3](<https://devfeed.tech/tags/amazon-s3.md>), [amazon-simple-storage-service-amazon-s3](<https://devfeed.tech/tags/amazon-simple-storage-service-amazon-s3.md>), [aws](<https://devfeed.tech/tags/aws.md>), [aws-support](<https://devfeed.tech/tags/aws-support.md>), [customer-solutions](<https://devfeed.tech/tags/customer-solutions.md>), [data](<https://devfeed.tech/tags/data.md>), [foundation](<https://devfeed.tech/tags/foundation.md>), [intermediate-200](<https://devfeed.tech/tags/intermediate-200.md>), [launch](<https://devfeed.tech/tags/launch.md>), [matching](<https://devfeed.tech/tags/matching.md>), [modernizing](<https://devfeed.tech/tags/modernizing.md>), [platform](<https://devfeed.tech/tags/platform.md>), [premium](<https://devfeed.tech/tags/premium.md>), [production](<https://devfeed.tech/tags/production.md>), [recruitment](<https://devfeed.tech/tags/recruitment.md>), [risk](<https://devfeed.tech/tags/risk.md>)

### AI overview

Thermo Fisher Scientific built Patient Match AI on a curated AWS patient recruitment data foundation that consolidates more than 20 million profiles (about 1.6 TB) from patient, survey, recruitment, marketing, and consent sources. Users can search in natural language and refine cohorts through a COUNT -> REFINE -> TARGETED LIST -> EXPORT workflow; shared test queries ran in under six seconds, and many common feasibility workflows took minutes rather than days. Aggregate views support exploration, while identity access remains subject to privacy, governance, consent, contactability, and PII controls. Central Insights Hub provides separate dashboards for recruitment progress, site and geographic performance, and other operational measures. AWS services include S3, Glue, Step Functions, Athena, Iceberg, DocumentDB, ECS, and Lambda. AWS Countdown Premium provided architecture and readiness reviews, launch support, and post-launch stabilization; its engineering guidance helped cut estimated processing time by more than threefold. The article reports a Q2 FY26 production launch.

### Source excerpt

Learn how Thermo Fisher Scientific built an AI-powered patient-matching platform on AWS, modernizing its clinical trial recruitment data foundation and using AWS Countdown Premium to de-risk the production launch.

## Beyond Symmetry: Asymmetric Matching with Transformers for Collective Entity Resolution

DevFeed: [Beyond Symmetry: Asymmetric Matching with Transformers for Collective Entity Resolution](<https://devfeed.tech/articles/beyond-symmetry-asymmetric-matching-with-transformers-for-collective-entity-resolution-132692.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.15185>)

Author: Andrei Olar

Published: 2026-11-13T05:00:00Z

Content type: opinion

Language: en

Sources: [arXiv: Databases](<https://devfeed.tech/sources/arxiv-databases.md>)

Topics: [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [Transformers](<https://devfeed.tech/topics/transformers.md>)

Tags: [collective](<https://devfeed.tech/tags/collective.md>), [cs-db](<https://devfeed.tech/tags/cs-db.md>), [entity-resolution](<https://devfeed.tech/tags/entity-resolution.md>), [matching](<https://devfeed.tech/tags/matching.md>), [symmetry](<https://devfeed.tech/tags/symmetry.md>), [transformers](<https://devfeed.tech/tags/transformers.md>)

### AI overview

The paper finds that input order can affect entity resolution matching and skew the final clusters. It studies a collective framework with separate matching and clustering stages, using pretrained transformers in an order-aware matcher to model asymmetry and measure its effects across benchmark datasets. The abstract reports that asymmetry is measurable and argues supervised matchers may need direction-aware training that preserves it, especially when information is vague or ambiguous. The supplied evidence does not give benchmark details or quantify the reported skew.

### Source excerpt

arXiv:2611.15185v1 Announce Type: new Abstract: A core assumption of entity resolution (ER) is its symmetric nature, which rests on a principle so self-evident it borders on tautology: if a refers to the same object as b, then b must also refer to the same object as a. However, this global property of symmetry does not necessarily extend to each of the complex decision processes within ER systems. In cases involving vague or ambiguous information, enforcing symmetry can be detrimental to performance, an issue exemplified by the ``bad triplet problem''. This paper investigates the effects of input-order asymmetry within a collective ER framework that integrates distinct matching and clustering stages. The input order-aware matching stage utilizes pre-trained transformers to explicitly model and exploit asymmetry, making it possible to evaluate its effect on clustering performance across multiple benchmark datasets. Our results demonstrate that asymmetry is not only measurable but also introduces a significant skew in the final entity clusters, highlighting the need to re-evaluate supervised training methodologies for matchers, particularly the development of direction-aware training paradigms that can preserve, rather than neutralize, the inherent asymmetry in matching tasks.

## DISCERN: Preserve at Write, Resolve at Read for Persistent LLM-Agent Memory

DevFeed: [DISCERN: Preserve at Write, Resolve at Read for Persistent LLM-Agent Memory](<https://devfeed.tech/articles/discern-preserve-at-write-resolve-at-read-for-persistent-llm-agent-memory-132165.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.15729>)

Author: Chirag Ahuja, Vladislav Tanov, Ramcharan Kakarla, Jonathan Taws

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [Retrieval Augmented Generation (RAG)](<https://devfeed.tech/topics/retrieval-augmented-generation-rag.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Enhancing AI model performance](<https://devfeed.tech/topics/enhancing-ai-model-performance.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-memory](<https://devfeed.tech/tags/agent-memory.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-loop](<https://devfeed.tech/tags/agentic-loop.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [enterprise-agents](<https://devfeed.tech/tags/enterprise-agents.md>), [entity-resolution](<https://devfeed.tech/tags/entity-resolution.md>), [framework](<https://devfeed.tech/tags/framework.md>), [graph](<https://devfeed.tech/tags/graph.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llm-as-a-judge](<https://devfeed.tech/tags/llm-as-a-judge.md>), [memory](<https://devfeed.tech/tags/memory.md>), [navigation](<https://devfeed.tech/tags/navigation.md>), [rag](<https://devfeed.tech/tags/rag.md>)

### AI overview

DISCERN targets persistent-agent memory resolution rather than retrieval: identifying the intended entity, the current time-state, and when to abstain. It preserves information in an additive bi-temporal provenance graph, keeping confusable entities separate and retaining superseded values; at read time, an agent navigates the graph through a zero-LLM tool surface and chooses among four gated terminal actions. On the 310-query synthetic ENT-310 benchmark, it scores 0.924 abstention F1 versus at most 0.802 for eight baselines and 0.861 macro answer correctness versus 0.806 for the strongest baseline, with the largest gain in entity resolution. With no tuning on the full LoCoMo corpus, it also leads macro correctness, 0.663 versus 0.638, and remains among the best calibrated on false-premise abstention. These results are reported in the supplied abstract.

### Source excerpt

arXiv:2611.15729v1 Announce Type: new Abstract: Enterprise agents accumulate memory across dozens of sessions in which facts contradict, get superseded, and blur as similarly named entities collide in embedding space. The field treats this as a retrieval problem; we show it is not. On ENT-310, a synthetic enterprise sales-agent memory benchmark of 310 queries (210 answerable, 100 unanswerable), a vanilla Retrieval-Augmented Generation (RAG) reader over a memory store already matches strong memory systems on the recall pillars. Retention and currency are near-saturated, roughly 0.93 to 0.98 for every competent system, so recall is essentially solved. What remains unsolved is resolution: which entity a query refers to, which time-state is current, and whether to commit an answer at all or decline when evidence is insufficient. We propose DISCERN (DISambiguation and Cross-session Episodic memoRy Navigation), an agent-memory framework built on a single rule: preserve at write, resolve at read. At write time DISCERN builds a strictly additive bi-temporal provenance graph that never merges confusable entities and never drops superseded values. At read time a zero-LLM tool surface drives an agentic loop that navigates this graph and commits exactly one of four gated terminal actions. A four-judge LLM-as-a-Judge council scores two axes: answer correctness on answerable questions, and the answer-versus-abstain decision as a classification problem. On ENT-310, DISCERN is the best-calibrated system on abstention (F1 0.924 versus at most 0.802 across eight baselines) and leads macro answer correctness under equal pillar weighting (0.861 versus 0.806 for the strongest baseline), a lead driven by entity resolution (0.800, where no baseline exceeds 0.40). Applied with zero tuning to the full LoCoMo corpus, the same pipeline again leads macro answer correctness (0.663 versus 0.638) and stays among the best-calibrated on false-premise abstention.

## PAL-Bench: Evidence-Grounded Profile Reconstruction from Longitudinal Personal Albums

DevFeed: [PAL-Bench: Evidence-Grounded Profile Reconstruction from Longitudinal Personal Albums](<https://devfeed.tech/articles/pal-bench-evidence-grounded-profile-reconstruction-from-longitudinal-personal-albums-131934.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2606.16175>)

Author: Qiwei Yan, Zhiqiang Yuan, Zexi Jia, Nanxing Hu, Kailin Lyu, Jie Zhou, Jinchao Zhang

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [Ground truth / benchmark quality](<https://devfeed.tech/topics/ground-truth-benchmark-quality.md>), [Audit-then-score](<https://devfeed.tech/topics/audit-then-score.md>), [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [Image processing](<https://devfeed.tech/topics/image-processing.md>), [sensitive data](<https://devfeed.tech/topics/sensitive-data.md>)

Tags: [album](<https://devfeed.tech/tags/album.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [best](<https://devfeed.tech/tags/best.md>), [builds](<https://devfeed.tech/tags/builds.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [data-integration](<https://devfeed.tech/tags/data-integration.md>), [entity-resolution](<https://devfeed.tech/tags/entity-resolution.md>), [framework](<https://devfeed.tech/tags/framework.md>), [identity](<https://devfeed.tech/tags/identity.md>), [metric](<https://devfeed.tech/tags/metric.md>), [privacy](<https://devfeed.tech/tags/privacy.md>), [profile](<https://devfeed.tech/tags/profile.md>), [provenance](<https://devfeed.tech/tags/provenance.md>), [public-records](<https://devfeed.tech/tags/public-records.md>), [records](<https://devfeed.tech/tags/records.md>), [social](<https://devfeed.tech/tags/social.md>), [synthetic](<https://devfeed.tech/tags/synthetic.md>), [temporal](<https://devfeed.tech/tags/temporal.md>), [timestamps](<https://devfeed.tech/tags/timestamps.md>)

### AI overview

PAL-Bench evaluates whether systems can reconstruct personal profiles and social relationships from noisy, longitudinal photo-album records while grounding claims in evidence. Its Evidence Compiler creates synthetic private worlds, renders album images, re-measures them through perception pipelines, and gives agents only perception-derived public records; targets, identity maps, and evidence paths stay hidden. The benchmark includes 50 synthetic users, 36,659 photo records, and 2,799 targets. An audit with 10 participants found its evidence structures match real albums, while equivalent public releases remain privacy-prohibitive. Across seven systems, results show that plausible owner summaries do not ensure accurate recurring-identity resolution or evidence citation. PAL-TRACE performs best by freezing identity bindings before mining owner facts, but difficult identity resolution remains unsolved.

### Source excerpt

arXiv:2606.16175v2 Announce Type: replace Abstract: Longitudinal personal albums are weak-schema multimodal databases: noisy perceptual records whose key facts require joins across faces, text, timestamps, locations, and repeated events. Existing visual, video, document, and lifelog benchmarks test sub-problems, but not album-scale profile reconstruction with social identity binding and evidence citation. Benchmarking this task is difficult because the ground truth needed for evaluation--owner profiles, social graphs, face-name maps, and evidence provenance--is private state that real albums cannot safely release. We introduce PAL-Bench, a controlled benchmark for evidence-grounded reconstruction under a public-record contract. Its Evidence Compiler builds latent private worlds, programs target-level evidence paths, renders album pixels, re-measures them through perception pipelines, and exports audited public/private views. Agents receive only perception-derived public records; targets, identifier maps, and evidence paths remain hidden. PAL-Bench contains 50 synthetic users, 36,659 public photo records, and 2,799 targets over owner facts, identities, and relations. A privacy-preserving audit with 10 participants confirms that PAL-Bench evidence structures match real private albums, though equivalent releases remain privacy-prohibitive. Across seven systems and two compute-matched diagnostics, a seven-metric protocol reveals a gap between plausible profile summarization and faithful social reconstruction: systems recover some owner facts but struggle with recurring identities and evidence citation. PAL-TRACE, a reference framework that freezes identity bindings before owner-fact mining, performs best but leaves hard identity resolution far from solved. PAL-Bench provides a testbed for perceptual entity resolution, multimodal data integration, temporal evidence aggregation, and provenance-aware structured prediction.

## How to share customer context across chat, SMS, and voice

DevFeed: [How to share customer context across chat, SMS, and voice](<https://devfeed.tech/articles/how-to-share-customer-context-across-chat-sms-and-voice-133765.md>)

Original publisher: [Read original article](<https://www.twilio.com/en-us/blog/insights/omnichannel-customer-context>)

Author: Jesse Sumrak

Published: 2026-11-13T00:00:00Z

Content type: tutorial

Language: en

Sources: [Twilio](<https://devfeed.tech/sources/twilio-blog.md>), [Twilio Blog](<https://devfeed.tech/sources/twilio-blog-2.md>)

Topics: [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [AI for customer service automation](<https://devfeed.tech/topics/ai-for-customer-service-automation.md>), [Conversational AI](<https://devfeed.tech/topics/conversational-ai.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [api](<https://devfeed.tech/tags/api.md>), [apis](<https://devfeed.tech/tags/apis.md>), [chat](<https://devfeed.tech/tags/chat.md>), [contact-center](<https://devfeed.tech/tags/contact-center.md>), [context](<https://devfeed.tech/tags/context.md>), [conversation-memory](<https://devfeed.tech/tags/conversation-memory.md>), [conversation-orchestrator](<https://devfeed.tech/tags/conversation-orchestrator.md>), [conversations](<https://devfeed.tech/tags/conversations.md>), [conversations-api](<https://devfeed.tech/tags/conversations-api.md>), [customer](<https://devfeed.tech/tags/customer.md>), [customers](<https://devfeed.tech/tags/customers.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [industry-insights](<https://devfeed.tech/tags/industry-insights.md>), [repeat](<https://devfeed.tech/tags/repeat.md>), [share](<https://devfeed.tech/tags/share.md>), [sms](<https://devfeed.tech/tags/sms.md>), [switch](<https://devfeed.tech/tags/switch.md>), [voice](<https://devfeed.tech/tags/voice.md>)

### AI overview

Twilio's approach depends on resolving every channel identifier to the same customer profile, then grouping conversations by that profile. Conversation Memory maps identifiers such as phone numbers, chat IDs, WhatsApp handles, and email to profiles; customers' alternate numbers must also be linked. Conversation Orchestrator uses profile grouping (GROUP_BY_PROFILE, which must be set explicitly) when resolution succeeds. Without a match, conversations fall back to grouping by address and do not connect to cross-channel history. Agent Connect can retrieve stored observations, summaries, and history before messages reach the agent, but its memory mode defaults to never and must be set to always or once. For chat-to-call handoffs that happen before a chat closes, save key facts as observations as they arise or use extraction on INACTIVE; otherwise the profile may not yet contain them. Human handoffs can pass profile and conversation IDs to Flex, where Unified Profiles shows customer history and Agent Copilot adds generated highlights; both are described as public beta.

### Source excerpt

Learn how to share customer context across chat, SMS, and voice, so customers don't repeat themselves when they switch channels.

## Towards a Privacy-Preserving Reliability-Aware Entity Resolution for Attributed Graphs \[Vision\] (Technical Report)

DevFeed: [Towards a Privacy-Preserving Reliability-Aware Entity Resolution for Attributed Graphs \[Vision\] (Technical Report)](<https://devfeed.tech/articles/towards-a-privacy-preserving-reliability-aware-entity-resolution-for-attributed-graphs-vision-technical-report-125662.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.10038>)

Author: Abdessamad Imine, Xiang Lian, Kambiz Ghazinour

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>), [arXiv: Databases](<https://devfeed.tech/sources/arxiv-databases.md>)

Topics: [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [Security & Privacy](<https://devfeed.tech/topics/security-privacy.md>)

Tags: [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-db](<https://devfeed.tech/tags/cs-db.md>), [entity-resolution](<https://devfeed.tech/tags/entity-resolution.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [privacy](<https://devfeed.tech/tags/privacy.md>), [reliability](<https://devfeed.tech/tags/reliability.md>), [vision](<https://devfeed.tech/tags/vision.md>)

### AI overview

Entity resolution for attributed graphs must balance accurate integration with privacy: reliability information helps account for differences in input quality, but retaining it in a published integrated graph can increase privacy risks. This research vision proposes a three-level coupled framework to unify reliability assessment and privacy preservation within the ER process, and identifies remaining challenges to publishing graphs that are both accurate and privacy-preserving. The supplied abstract does not specify the framework's levels or implementation details.

### Source excerpt

arXiv:2611.10038v1 Announce Type: cross Abstract: Entity resolution (ER) on attributed graphs aims to identify nodes across multiple graphs that correspond to the same real-world entity and merge them into an integrated graph. When input graphs vary in quality and contain sensitive information, ER faces a dual challenge: while reliability information is crucial for accurate graph data integration, its presence in the published integrated graph can heighten privacy risks. This paper presents a research vision that unifies reliability assessment and privacy preservation within a single graph ER process. We highlight the inherent tension between these two objectives, propose a three-level coupled framework, and outline the remaining research challenges to enable the publication of integrated graphs that are both accurate and privacy-preserving.

## Why Your LLM Keeps Matching the Wrong Products (And Why a Better Prompt Won't Fix It)

DevFeed: [Why Your LLM Keeps Matching the Wrong Products (And Why a Better Prompt Won't Fix It)](<https://devfeed.tech/articles/why-your-llm-keeps-matching-the-wrong-products-and-why-a-better-prompt-won-t-fix-it-114862.md>)

Original publisher: [Read original article](<https://hackernoon.com/why-your-llm-keeps-matching-the-wrong-products-and-why-a-better-prompt-wont-fix-it?source=rss>)

Author: Oleg Kokorin

Published: 2026-11-03T11:25:53Z

Content type: article

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [AI-driven information retrieval](<https://devfeed.tech/topics/ai-driven-information-retrieval.md>), [business logic](<https://devfeed.tech/topics/business-logic.md>), [Enhancing AI model performance](<https://devfeed.tech/topics/enhancing-ai-model-performance.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>)

Tags: [abbreviation](<https://devfeed.tech/tags/abbreviation.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [algorithm](<https://devfeed.tech/tags/algorithm.md>), [algorithms](<https://devfeed.tech/tags/algorithms.md>), [alternatives](<https://devfeed.tech/tags/alternatives.md>), [approach](<https://devfeed.tech/tags/approach.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [automate](<https://devfeed.tech/tags/automate.md>), [automated](<https://devfeed.tech/tags/automated.md>), [business](<https://devfeed.tech/tags/business.md>), [candidate](<https://devfeed.tech/tags/candidate.md>), [catalog-matching](<https://devfeed.tech/tags/catalog-matching.md>), [facets-of-likelihood](<https://devfeed.tech/tags/facets-of-likelihood.md>), [fine-tuning-llms](<https://devfeed.tech/tags/fine-tuning-llms.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llm-prompting](<https://devfeed.tech/tags/llm-prompting.md>), [llms](<https://devfeed.tech/tags/llms.md>), [matching](<https://devfeed.tech/tags/matching.md>), [product-catalog-ai-prompts](<https://devfeed.tech/tags/product-catalog-ai-prompts.md>), [product-matching](<https://devfeed.tech/tags/product-matching.md>), [products](<https://devfeed.tech/tags/products.md>), [prompt](<https://devfeed.tech/tags/prompt.md>), [rag-architecture](<https://devfeed.tech/tags/rag-architecture.md>), [rules](<https://devfeed.tech/tags/rules.md>), [search](<https://devfeed.tech/tags/search.md>), [similarity](<https://devfeed.tech/tags/similarity.md>), [text](<https://devfeed.tech/tags/text.md>)

### AI overview

Product matching errors often come from retrieval, missing data, attribute handling, or business rules rather than prompt quality. The article recommends a staged system: retrieve candidates with lexical and semantic search, extract and normalize relevant attributes, then rank pairs using structured features and category specific rules. Evaluate candidate retrieval separately, since a correct product omitted from the candidate list cannot be selected later. Use configuration for changing rules and synonyms, and retrieve validated historical examples for difficult cases; irrelevant or outdated examples can mislead. Before changing models, check candidate coverage, input completeness, attribute normalization, and rule application. Applying LLMs only to retrieved candidates can lower processing costs but risks missing valid matches; broader processing may improve recall at higher cost. Cloud models have usage fees, while local inference requires infrastructure and maintenance, so compare total costs at the required quality.

### Source excerpt

Why do LLMs match the wrong products? Learn why text similarity falls short and how candidate search, attribute checks, and business rules improve matching.

## Rebuilding a Field's Open-Problem Catalog from the Literature: A Two-Layer Knowledge Graph, and Why Status Must Stay Proposed

DevFeed: [Rebuilding a Field's Open-Problem Catalog from the Literature: A Two-Layer Knowledge Graph, and Why Status Must Stay Proposed](<https://devfeed.tech/articles/rebuilding-a-field-s-open-problem-catalog-from-the-literature-a-two-layer-knowledge-graph-and-why-status-must-stay-proposed-108220.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.39641>)

Author: Madhulatha Mandarapu, Sandeep Kunkunuru

Published: 2026-10-30T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Databases](<https://devfeed.tech/sources/arxiv-databases.md>)

Topics: [graph-database](<https://devfeed.tech/topics/graph-database.md>), [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-research](<https://devfeed.tech/tags/ai-research.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [audit](<https://devfeed.tech/tags/audit.md>), [build](<https://devfeed.tech/tags/build.md>), [can](<https://devfeed.tech/tags/can.md>), [candidate](<https://devfeed.tech/tags/candidate.md>), [catalog](<https://devfeed.tech/tags/catalog.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-db](<https://devfeed.tech/tags/cs-db.md>), [cs-dl](<https://devfeed.tech/tags/cs-dl.md>), [field](<https://devfeed.tech/tags/field.md>), [graph](<https://devfeed.tech/tags/graph.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [knowledge-graph](<https://devfeed.tech/tags/knowledge-graph.md>), [layer](<https://devfeed.tech/tags/layer.md>), [literature](<https://devfeed.tech/tags/literature.md>), [open](<https://devfeed.tech/tags/open.md>), [status](<https://devfeed.tech/tags/status.md>), [two-layer](<https://devfeed.tech/tags/two-layer.md>)

### AI overview

The paper builds a queryable open-problem catalog by keeping a hand-curated problem layer intact and joining it to a bibliographic evidence layer that is rebuilt, using typed links with provenance. Its language model extracts paper details, grounds them to catalog entries, proposes per-paper status verdicts, and generates ideas; each task is limited to candidate sets so unsupported matches are rejected. On catalogs of 1,053 database problems and 1,000 AI problems, the graph retained nearly all prior catalog-only question coverage (47/47 and 34/35) and enabled queries about recent workers, untouched problems, methods, and related outside papers. Status judgments were unreliable: of 41 checked solve/refute verdicts, 9 were correct, 30 partly correct, and 2 wrong; none of 34 refutes disproved an open problem, and 1 of 7 solves closed the central question. The authors therefore require status changes to remain evidence-backed proposals for human review, and identify entity resolution, verification, source-quota freshness, and evaluation without ground truth as remaining challenges.

### Source excerpt

arXiv:2610.39641v1 Announce Type: cross Abstract: A research field records its open problems in surveys, keynotes and community reports. Those are prose, written every few years, and nobody can ask them who works on a problem now or whether it was settled last year. We argue the problems should be data, and we build the system that makes them so: a hand-curated problem layer carried in unchanged, under an evidence layer rebuilt from a bibliographic index, joined by typed, provenance-carrying links in one graph database and queryable from one language. A language model does four jobs -- per-paper extraction, catalog grounding, per-paper status verdict, and idea generation -- each constrained to a candidate set so out-of-set answers are refused rather than stored. We report a study on two curated catalogs, database research (1,053 problems) and AI research (1,000 problems), each with a prior catalog-only graph. The rebuilt graph answers every question the catalog-only graph answered -- 47 of 47 on databases, 34 of 35 on AI -- plus questions the catalog-only graph refused: recent workers, untouched problems, methods tried, and outside papers bearing on a problem (336 problems, 770 works on databases). The capability we most wanted does not survive its own audit. Of 41 solves and refutes verdicts hand-checked against the papers, 9 are correct, 30 partly correct and 2 wrong; no refutes verdict (0 of 34) disproves an open problem, and only 1 of 7 solves closes the central question. We report that failure in full and draw the design conclusion: a status change must be proposed with its evidence and never applied, making the human reviewer part of the architecture. We close with the data-engineering problems this raises: entity resolution across curated and harvested layers, verifying model status claims, freshness under source quotas, and evaluation without ground truth.

## Constraining the Commit Boundary: A DIO/ODI-Informed Authorization Guardrail for Agentic Entity Resolution

DevFeed: [Constraining the Commit Boundary: A DIO/ODI-Informed Authorization Guardrail for Agentic Entity Resolution](<https://devfeed.tech/articles/constraining-the-commit-boundary-a-dio-odi-informed-authorization-guardrail-for-agentic-entity-resolution-103267.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.37407>)

Author: Daniel Ber\\'anek

Published: 2026-10-28T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>), [arXiv: Databases](<https://devfeed.tech/sources/arxiv-databases.md>)

Topics: [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [Authorization](<https://devfeed.tech/topics/authorization.md>), [Embabel](<https://devfeed.tech/topics/embabel.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [actions](<https://devfeed.tech/tags/actions.md>), [admission](<https://devfeed.tech/tags/admission.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [authorization](<https://devfeed.tech/tags/authorization.md>), [boundary](<https://devfeed.tech/tags/boundary.md>), [commit](<https://devfeed.tech/tags/commit.md>), [condition](<https://devfeed.tech/tags/condition.md>), [constraint](<https://devfeed.tech/tags/constraint.md>), [contribution](<https://devfeed.tech/tags/contribution.md>), [coverage](<https://devfeed.tech/tags/coverage.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-db](<https://devfeed.tech/tags/cs-db.md>), [dependency](<https://devfeed.tech/tags/dependency.md>), [digital](<https://devfeed.tech/tags/digital.md>), [digital-identity](<https://devfeed.tech/tags/digital-identity.md>), [domain](<https://devfeed.tech/tags/domain.md>), [entity-resolution](<https://devfeed.tech/tags/entity-resolution.md>), [guardrail](<https://devfeed.tech/tags/guardrail.md>)

### AI overview

The paper proposes a DIO/ODI-informed authorization boundary for agentic entity resolution that separates evidence an agent may use for reasoning from evidence allowed to authorize persistent identity changes. A two-phase supervisor permits a commit only when claim-specific eligible evidence covers every required proposition, material conflicts are resolved, the operation stays within its authorized scope, and relevant state and policy versions remain valid. UNVERIFIED admission alone does not make evidence eligible; insufficient support or unresolved conflict leads to deferral, with separate claim-specific promotion controlling evidentiary use. The proposed evaluation would test whether mapping reconstruction-failure distinctions to identity-preserving constraints adds value beyond generic authorization and identifier rules. This is a conceptual specification without implementation or empirical validation; it depends on adequate domain policy, evidence promotion, and enforcement. It does not prevent mistaken hypotheses or hypothesis-conditioned retrieval, which remain outside the proposed boundary.

### Source excerpt

arXiv:2610.37407v1 Announce Type: new Abstract: Agentic Entity Resolution (ER) systems iteratively plan actions, retrieve evidence, revise hypotheses, and propose graph mutations. An erroneous identity hypothesis may condition later retrieval and produce apparent confirmation. This paper proposes a Digital Identity Optimization (DIO) and Ontology of Digital Identity (ODI)-informed authorization boundary between evidence available for agent reasoning and evidence eligible to authorize persistent identity-changing mutations. A domain-governed constraint-binding function maps selected reconstruction-failure distinctions to explicit identity-preserving constraints. An abstract two-phase supervisor permits a commit only when claim-specific authorization-eligible evidence supports every required proposition, relevant conflicts are resolved, the operation remains within its authorized scope, and relevant state and policy versions remain valid. Admission as UNVERIFIED does not confer authorization eligibility. Separate claim-specific promotion governs evidentiary use; insufficient support or unresolved material conflict leads to deferral. The contribution is a specification connecting diagnostic-to-constraint mapping, evidence promotion, proposition coverage, conflict handling, scope binding, and commit-time version integrity. A proposed evaluation protocol is designed to assess the mapping's incremental value beyond generic authorization and conventional identifier rules. This is a conceptual architecture without implementation or empirical validation, not a theorem of semantic correctness. Its property depends on domain-policy adequacy, evidence promotion, and enforcement. It does not prevent erroneous hypotheses or hypothesis-conditioned retrieval; dependency-aware evidence flow remains future research.

## ActionRAG: Evidence-Refining Graph Diffusion for Long-Term Memory in LLM Agents

DevFeed: [ActionRAG: Evidence-Refining Graph Diffusion for Long-Term Memory in LLM Agents](<https://devfeed.tech/articles/actionrag-evidence-refining-graph-diffusion-for-long-term-memory-in-llm-agents-100042.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.32752>)

Author: Navyansh Mahla, Sashank Vandrangi

Published: 2026-10-26T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Graphs](<https://devfeed.tech/topics/graphs.md>), [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [diffusion](<https://devfeed.tech/tags/diffusion.md>), [entity-resolution](<https://devfeed.tech/tags/entity-resolution.md>), [flux](<https://devfeed.tech/tags/flux.md>), [graph](<https://devfeed.tech/tags/graph.md>), [indexing](<https://devfeed.tech/tags/indexing.md>), [learning](<https://devfeed.tech/tags/learning.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llm-agents](<https://devfeed.tech/tags/llm-agents.md>), [materialization](<https://devfeed.tech/tags/materialization.md>), [memory](<https://devfeed.tech/tags/memory.md>), [memory-architecture](<https://devfeed.tech/tags/memory-architecture.md>), [merge](<https://devfeed.tech/tags/merge.md>), [online](<https://devfeed.tech/tags/online.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [pipelines](<https://devfeed.tech/tags/pipelines.md>), [points](<https://devfeed.tech/tags/points.md>), [protection](<https://devfeed.tech/tags/protection.md>), [query](<https://devfeed.tech/tags/query.md>), [retrieval](<https://devfeed.tech/tags/retrieval.md>)

### AI overview

ActionRAG combines cautious graph indexing with query-time retrieval refinement to recover consistent evidence from an LLM agent's growing interaction history. It reconciles type labels and uses tiered entity resolution, then retrieves entities and triples to build a bounded local graph. Query-conditioned diffusion adjusts triple scores using local paths, while seed protection and hub-penalized flux preserve direct evidence and help recover connections. On MemoryAgentBench, it reports 55.1 average accuracy across eight deterministic-scoring columns, 5.4 points above the strongest retrieval-based memory baseline. Mean synthesis cost is below $0.001 per query on four reported cost columns; separately measured retrieval-side p99 components sum to 440.7 ms on a consumer Apple Silicon setup.

### Source excerpt

arXiv:2610.32752v1 Announce Type: new Abstract: LLM agents need long-term memory that can recover precise and mutually consistent evidence from growing interaction histories, particularly when answers depend on relations among facts rather than isolated semantic matches. Current memory pipelines remain vulnerable at two stages: indexing errors can fragment repeated entities or merge distinct ones, and online retrieval often treats initial similarity scores as final even when the surrounding graph provides evidence for revising them. We introduce ActionRAG, a graph memory architecture that couples cautious indexing with query-time graph retroevaluation. Its indexing pipeline reconciles type labels and applies tiered entity resolution. Given a query, dual-channel entity and triple retrieval forms an initial evidence set and induces a bounded local graph. Query-conditioned receiver-normalized diffusion refines triple scores using local path structure. Seed protection, hub-penalized flux, and triple materialization preserve direct evidence while recovering missed connections. We evaluate ActionRAG on MemoryAgentBench (MAB), which tests accurate retrieval, conflict resolution, long-range understanding, and test-time learning. Across its eight deterministic-scoring columns, ActionRAG achieves 55.1 average accuracy, improving over the strongest retrieval-based memory baseline by 5.4 points. Under MAB's query-time cost accounting, mean synthesis cost stays below \$0.001 per query on the four reported cost columns, and the sum of separately measured retrieval-side p99 components is 440.7 ms on our consumer Apple Silicon setup.

## Ontology-Guided Knowledge Graph Construction from Structured Metadata and Climate Science Publications

DevFeed: [Ontology-Guided Knowledge Graph Construction from Structured Metadata and Climate Science Publications](<https://devfeed.tech/articles/ontology-guided-knowledge-graph-construction-from-structured-metadata-and-climate-science-publications-100037.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.32725>)

Author: Aayush Acharya, Qi Zhang, Eduard Dragut, Longin Jan Latecki

Published: 2026-10-26T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [база знаний](<https://devfeed.tech/topics/tag-c51cc4bf76e5.md>), [knowledge-engineering](<https://devfeed.tech/topics/knowledge-engineering.md>), [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [Structured-data](<https://devfeed.tech/topics/structured-data.md>), [Data Science](<https://devfeed.tech/topics/data-science.md>)

Tags: [climate](<https://devfeed.tech/tags/climate.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [knowledge-graph](<https://devfeed.tech/tags/knowledge-graph.md>)

### AI overview

The framework turns climate science publications into an ontology-grounded knowledge graph by having an LLM extract evidence-backed triples, then using specialized agents and multistage checks for ontology conformance and entity linking. Schema constraints and quoted text fragments are intended to make entries consistent and traceable. In an agreement study of over 1,000 triples, Gemini-2.5-flash achieved Cohen's κ of 0.640-0.700 against three annotators, within the reported human-to-human range of 0.616-0.689. Applied to 2,000 publications, the pipeline produced a graph with 133,990 nodes and 755,308 relations; 98.96% of 23,327 validated publication-contributed triples were judged correct. These are reported evaluation results, not a guarantee of accuracy for other corpora.

### Source excerpt

arXiv:2610.32725v1 Announce Type: cross Abstract: The rapid expansion of scientific literature far outpaces the speed of manual curation, creating a fundamental bottleneck in extracting structured knowledge from scientific texts. To address this, we introduce an agentic, ontology-guided framework for knowledge graph (KG) construction. Our framework utilizes specialized agents for document understanding, information extraction, ontology conformance, and entity linking. First, an LLM extracts factual knowledge in the form of triples with explicit textual evidence from research papers. Then the triplets undergo a multistage verification process to ground the triplets in the ontology. By enforcing schema constraints and requiring evidence in the form of text fragments, we ensure the data is both semantically consistent and traceable. In an agreement study over 1,000 extracted triplets, Gemini-2.5-flash achieves Cohen's $\kappa$ scores of 0.640--0.700 against three human annotators, matching the human--human agreement range of 0.616--0.689. We applied this framework to 2,000 scientific publications and construct a novel, ontology-grounded KG for climate science with 133,990 nodes and 755,308 relations; of the 23,327 triplets that the publications contribute after validation, 98.96\% are judged correct. The pipeline code and KG is publicly available https://github.com/aayushacharya/climatekg.

## Same Entity, Different Script: Diagnosing Cross-Script Entity Matching Failures in Language Models

DevFeed: [Same Entity, Different Script: Diagnosing Cross-Script Entity Matching Failures in Language Models](<https://devfeed.tech/articles/same-entity-different-script-diagnosing-cross-script-entity-matching-failures-in-language-models-98212.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.31407>)

Author: Athira Gopal, Patil Prathamesh Pravin, Ankush Kadu, Arpit Jain, Ashwanth Krishnan

Published: 2026-10-23T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [qwen](<https://devfeed.tech/topics/qwen.md>), [Internationalization (i18n)](<https://devfeed.tech/topics/i18n.md>), [enhancing AI model accuracy](<https://devfeed.tech/topics/enhancing-ai-model-accuracy.md>)

Tags: [activation](<https://devfeed.tech/tags/activation.md>), [cross](<https://devfeed.tech/tags/cross.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [discrimination](<https://devfeed.tech/tags/discrimination.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [language](<https://devfeed.tech/tags/language.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [matching](<https://devfeed.tech/tags/matching.md>), [script](<https://devfeed.tech/tags/script.md>), [study](<https://devfeed.tech/tags/study.md>), [tokenization](<https://devfeed.tech/tags/tokenization.md>)

### AI overview

On audited name pairs, language models often fail to judge two script variants as the same person, even when they usually identify different people correctly. For Qwen2.5-7B, same-person accuracy fell from 35.1% with Latin names to 8.5% with Devanagari under Marathi prompts; a second evaluation found 5.3% for Malayalam. Tests across five models found distinct response biases, and reducing tokenization fragmentation did not reliably eliminate the gap. Probes showed that Devanagari model states retained identity information, while scalar calibration recovered 87% of the gap between raw model and probe balanced accuracy, implicating decision criteria as a major factor. Activation patching localized causal sensitivity to the second name in Qwen's middle layers. A fact-independent intervention there raised held-out Marathi same-person accuracy from 6.5% to 33.5%, with 98.5% different-person accuracy. These findings are based on the study's audited pairs and reported model evaluations.

### Source excerpt

arXiv:2610.31407v1 Announce Type: cross Abstract: Multilingual language models can give different identity judgments when the same people's names are written in different scripts. We study this effect using 601 audited identity pairs with attested Devanagari and Latin forms, holding the question and relation fixed. Qwen2.5-7B's same-person accuracy falls from 35.1% in Latin to 8.5% in Devanagari under Marathi prompts, while different-person accuracy remains above 98%. Evaluation across five models reveals distinct response biases, and reduced tokenization fragmentation does not reliably close the gap. Replicating the design on 374 audited Malayalam pairs built from the same entities reproduces the effect in a second script and language family: same-person accuracy falls to 5.3%, about half the matched Devanagari rate, while the margin still ranks matching above non-matching names at 0.925 AUROC. Latin-trained probes recover label information from Devanagari states, exceeding a string-feature baseline under similarity-matched negatives. Scalar calibration recovers 87% of the raw model-probe balanced-accuracy gap, identifying decision criterion placement as a major contributor. Activation patching localises causal sensitivity to the second entity in Qwen's middle layers; one shared sensitivity direction transfers about half the full-state margin correction. A fact-independent intervention at this site raises held-out Marathi same-person accuracy from 6.5% to 33.5%, with 98.5% different-person accuracy. Together, these results distinguish script-dependent discrimination from criterion placement and identify a causal site for improving identity judgments.

## Cross-store product clustering across millions of stores

DevFeed: [Cross-store product clustering across millions of stores](<https://devfeed.tech/articles/cross-store-product-clustering-across-millions-of-stores-90670.md>)

Original publisher: [Read original article](<https://shopify.engineering/catalog-clustering-part-ii>)

Author: Ilia Kuchumov

Published: 2026-10-20T13:48:59Z

Content type: article

Language: en

Sources: [Shopify Engineering](<https://devfeed.tech/sources/shopify-engineering.md>), [Shopify Engineering - Shopify Engineering](<https://devfeed.tech/sources/shopify-engineering-shopify-engineering.md>)

Topics: [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [clustering](<https://devfeed.tech/topics/clustering.md>), [resource orchestration](<https://devfeed.tech/topics/resource-orchestration.md>), [AI Models](<https://devfeed.tech/topics/ai-models.md>), [product mindset](<https://devfeed.tech/topics/product-mindset.md>), [distributed-training](<https://devfeed.tech/topics/distributed-training.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [algorithm](<https://devfeed.tech/tags/algorithm.md>), [api](<https://devfeed.tech/tags/api.md>), [approximate-nearest-neighbor](<https://devfeed.tech/tags/approximate-nearest-neighbor.md>), [approximate-nearest-neighbor-search](<https://devfeed.tech/tags/approximate-nearest-neighbor-search.md>), [based](<https://devfeed.tech/tags/based.md>), [cluster](<https://devfeed.tech/tags/cluster.md>), [clustering](<https://devfeed.tech/tags/clustering.md>), [clusters](<https://devfeed.tech/tags/clusters.md>), [graph](<https://devfeed.tech/tags/graph.md>), [has](<https://devfeed.tech/tags/has.md>), [matching](<https://devfeed.tech/tags/matching.md>), [partitioning](<https://devfeed.tech/tags/partitioning.md>)

### AI overview

Shopify describes a system that matches product listings across merchants and groups likely matches under a unified product identifier. It combines approximate-neighbor candidate retrieval, tiered matching models, graph partitioning, and learned merge decisions, while prioritizing precision and refreshing clusters as listings change.

### Source excerpt

How we group product listings across millions of stores using tiered matching models, graph partitioning, and learned cluster merge decisions.

## The Technology Behind the Names: How We're Bringing Local War Memorials to Life

DevFeed: [The Technology Behind the Names: How We're Bringing Local War Memorials to Life](<https://devfeed.tech/articles/the-technology-behind-the-names-how-we-re-bringing-local-war-memorials-to-life-92459.md>)

Original publisher: [Read original article](<https://tech.findmypast.com/technology-behind-names/>)

Author: Sophie Adair

Published: 2026-10-20T13:00:00Z

Content type: article

Language: en

Sources: [Findmypast](<https://devfeed.tech/sources/findmypast.md>)

Topics: [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [RDF (Resource Description Framework)](<https://devfeed.tech/topics/rdf.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [Image](<https://devfeed.tech/topics/image.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [ai-model](<https://devfeed.tech/tags/ai-model.md>), [applied-ai](<https://devfeed.tech/tags/applied-ai.md>), [archive](<https://devfeed.tech/tags/archive.md>), [data](<https://devfeed.tech/tags/data.md>), [data-scientists](<https://devfeed.tech/tags/data-scientists.md>), [engineer](<https://devfeed.tech/tags/engineer.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [family](<https://devfeed.tech/tags/family.md>), [findmypast](<https://devfeed.tech/tags/findmypast.md>), [linking](<https://devfeed.tech/tags/linking.md>), [links](<https://devfeed.tech/tags/links.md>), [local](<https://devfeed.tech/tags/local.md>), [military](<https://devfeed.tech/tags/military.md>), [records](<https://devfeed.tech/tags/records.md>), [search](<https://devfeed.tech/tags/search.md>), [stories](<https://devfeed.tech/tags/stories.md>), [technology](<https://devfeed.tech/tags/technology.md>), [war](<https://devfeed.tech/tags/war.md>)

### AI overview

Findmypast describes a project linking First World War soldiers to local memorials and reconstructing their lives from records and newspapers. It connects Soldiers Who Died in the Great War and Commonwealth War Graves Commission transcripts by name and service number, then uses birth details and CWGC family notes to link soldiers to census records and build a persistent graph of people and relationships. Embeddings help handle spelling variations, related job terms, and aliases. AI-generated newspaper searches are checked against each soldier's details; memorial matches are prioritized using birthplace, census addresses, next-of-kin locations, and newspaper name lists. Because no single source of truth exists, the project does not claim perfect memorial matching. Narratives are generated only from sourced profile facts, with uncertainty stated, and were compared with hand-researched cases. The article reports more than 9,000 memorials and 300,000 men's stories.

### Source excerpt

The Technology Behind the Names: How We're Bringing Local War Memorials to Life In a recent survey, 73% of respondents said they believe the stories behind the names on local war memorials are getting lost, and only 27% were confident they could name even one person commemorated on their local memorial. We wanted to change that and bring people the stories behind the names on their local WW1 memorials. At scale, this is no small feat. What we know about any one soldier is scattered across dozens of record sets, both military and civilian. But with Findmypast's records and newspaper archive, some carefully applied AI, and a brilliant group of engineers and data scientists, we've begun a multi-year project to do exactly that. Here's how it works. Step 1. Finding the soldiers One of the things that has never been done before at this scale is reliably link everything we know about a soldier. Russell Hunter (Engineer Mavericks team) started with two key record sets, both available on Findmypast: Soldiers Who Died in the Great War, and the Commonwealth War Graves Commission transcripts. By linking the two by name and service number we could establish the basic facts for each fallen soldier: place and date of birth, regiment, and place and date of death. From there it was a relatively short hop to medal index cards and surviving service records, giving us a clear picture of each man's service during the war. Step 2. Finding their civilian lives When it comes to finding out more about their civilian lives, we really wanted to create links from the military records to the 1891/1901/1911 census records. Birth date and place help to make these links, but the most valuable clue came from the CWGC family notes, which for many soldiers record the names and address of parents or a spouse. A search for James Bourner, born 1888, might return dozens of matches. A search for James Bourner, son of Mrs C Bourner, of Forge Square Leigh is going to narrow that down hugely. Once we were ab

## Blocking Identity Resolution Infrastructure

DevFeed: [Blocking Identity Resolution Infrastructure](<https://devfeed.tech/articles/blocking-identity-resolution-infrastructure-90028.md>)

Original publisher: [Read original article](<https://webkit.org/blog/18438/blocking-identity-resolution-infrastructure/>)

Author: John Wilander; Charlie Wolfe; Matthew Finkel

Published: 2026-10-19T17:50:24Z

Content type: news

Language: en

Sources: [WebKit](<https://devfeed.tech/sources/webkit.md>)

Topics: [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [pii](<https://devfeed.tech/topics/pii.md>), [WebKit](<https://devfeed.tech/topics/webkit.md>), [frontend security](<https://devfeed.tech/topics/frontend-security.md>)

Tags: [identity](<https://devfeed.tech/tags/identity.md>), [privacy](<https://devfeed.tech/tags/privacy.md>), [tracking](<https://devfeed.tech/tags/tracking.md>), [web](<https://devfeed.tech/tags/web.md>), [webkit](<https://devfeed.tech/tags/webkit.md>)

### AI overview

WebKit describes identity resolution as a post-cookie tracking method in which third parties collect users' email addresses and link them to persistent tracking IDs and identity graphs. It says Safari and WebKit in iOS, iPadOS, macOS, and visionOS 27 block network access to this infrastructure at the domain level, while allowing first-party personalization.

### Source excerpt

Apple and WebKit have shipped protections against web tracking for a long time.

## Google Maps Bookings Have an Attribution Problem. Can Your Data Pipeline Explain Why?

DevFeed: [Google Maps Bookings Have an Attribution Problem. Can Your Data Pipeline Explain Why?](<https://devfeed.tech/articles/google-maps-bookings-have-an-attribution-problem-can-your-data-pipeline-explain-why-90142.md>)

Original publisher: [Read original article](<https://hackernoon.com/google-maps-bookings-have-an-attribution-problem-can-your-data-pipeline-explain-why?source=rss>)

Author: Mayur S

Published: 2026-10-19T11:21:59Z

Content type: tutorial

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [Entity resolution](<https://devfeed.tech/topics/entity-resolution.md>), [Google Analytics](<https://devfeed.tech/topics/google-analytics.md>), [AdMob](<https://devfeed.tech/topics/admob.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [crm](<https://devfeed.tech/tags/crm.md>), [crm-integration](<https://devfeed.tech/tags/crm-integration.md>), [data](<https://devfeed.tech/tags/data.md>), [data-architecture](<https://devfeed.tech/tags/data-architecture.md>), [data-pipeline](<https://devfeed.tech/tags/data-pipeline.md>), [google-ads](<https://devfeed.tech/tags/google-ads.md>), [google-local-service-ads](<https://devfeed.tech/tags/google-local-service-ads.md>), [google-maps-attribution](<https://devfeed.tech/tags/google-maps-attribution.md>), [lead-tracking](<https://devfeed.tech/tags/lead-tracking.md>), [local-service-ads](<https://devfeed.tech/tags/local-service-ads.md>), [october-2026-lsa-update](<https://devfeed.tech/tags/october-2026-lsa-update.md>), [performance-max](<https://devfeed.tech/tags/performance-max.md>)

### AI overview

The article explains how agencies can track bookings that begin through Google Maps and distinguish advertising leads from organic bookings and business revenue. It recommends preserving separate identifiers and event relationships across booking, CRM, advertising, and invoicing systems, recording evidence and status changes in an event ledger, and monitoring mismatches. It also stresses that Google's reported billing status does not establish whether an appointment was completed or generated revenue.

### Source excerpt

Google Maps booking attribution connects ad interactions, bookings, CRM records, and revenue. Learn how to track booking sources.

[Next page](<https://devfeed.tech/topics/entity-resolution.md?cursor=WyIyMDI2LTEwLTE5VDExOjIxOjU5KzAwOjAwIiwgIjI3MjBjNjFjLTdmNjItNDA5Zi05ZjNjLWZiZTRjZmEyZWY2OSJd>)