A plethora of Entity Linking (EL) approaches has recently been developed. While many claim to be multilingual, the MAG (Multilingual AGDISTIS) approach has been shown recently to outperform the state of the art in multilingual EL on 7 languages. With this demo, we extend MAG to support EL in 40 different languages, including especially low-resources languages such as Ukrainian, Greek, Hungarian, Croatian, Portuguese, Japanese and Korean. Our demo relies on online web services which allow for an easy access to our entity linking approaches and can disambiguate against DBpedia and Wikidata. During the demo, we will show how to use MAG by means of POST requests as well as using its user-friendly web interface. All data used in the demo is available at https://hobbitdata.informatik.uni-leipzig.de/agdistis/
We examine the problem of question answering over knowledge graphs, focusing on simple questions that can be answered by the lookup of a single fact. Adopting a straightforward decomposition of the problem into entity detection, entity linking, relation prediction, and evidence combination, we explore simple yet strong baselines. On the popular SimpleQuestions dataset, we find that basic LSTMs and GRUs plus a few heuristics yield accuracies that approach the state of the art, and techniques that do not use neural networks also perform reasonably well. These results show that gains from sophisticated deep learning techniques proposed in the literature are quite modest and that some previous models exhibit unnecessary complexity.
A major proportion of a text summary includes important entities found in the original text. These entities build up the topic of the summary. Moreover, they hold commonsense information once they are linked to a knowledge base. Based on these observations, this paper investigates the usage of linked entities to guide the decoder of a neural text summarizer to generate concise and better summaries. To this end, we leverage on an off-the-shelf entity linking system (ELS) to extract linked entities and propose Entity2Topic (E2T), a module easily attachable to a sequence-to-sequence model that transforms a list of entities into a vector representation of the topic of the summary. Current available ELS's are still not sufficiently effective, possibly introducing unresolved ambiguities and irrelevant entities. We resolve the imperfections of the ELS by (a) encoding entities with selective disambiguation, and (b) pooling entity vectors using firm attention. By applying E2T to a simple sequence-to-sequence model with attention mechanism as base model, we see significant improvements of the performance in the Gigaword (sentence to title) and CNN (long document to multi-sentence highlights) summarization datasets by at least 2 ROUGE points.
Many question answering systems over knowledge graphs rely on entity and relation linking components in order to connect the natural language input to the underlying knowledge graph. Traditionally, entity linking and relation linking have been performed either as dependent sequential tasks or as independent parallel tasks. In this paper, we propose a framework called EARL, which performs entity linking and relation linking as a joint task. EARL implements two different solution strategies for which we provide a comparative analysis in this paper: The first strategy is a formalisation of the joint entity and relation linking tasks as an instance of the Generalised Travelling Salesman Problem (GTSP). In order to be computationally feasible, we employ approximate GTSP solvers. The second strategy uses machine learning in order to exploit the connection density between nodes in the knowledge graph. It relies on three base features and re-ranking steps in order to predict entities and relations. We compare the strategies and evaluate them on a dataset with 5000 questions. Both strategies significantly outperform the current state-of-the-art approaches for entity and relation linking.
대규모 비정형 뉴스 코퍼스에서 정치 엘리트의 비공식·적대적 관계를 추출하기 위해, 스팬 기반 개체명 인식(NER), Wikidata 3단계 연결, 온톨로지 제약 전문가 혼합(MoE) 모델의 가이드 디코딩을 결합한 완전 공개 가중치 다국어 공동 개체-관계 추출 파이프라인을 제안한다. 3,491개 관계 골드 스탠다드 전수 점검에서 엄격 기준 68.2%·완화 기준 93.7%의 텍스트 정확도를 보였고, 오스트리아 정당 생애주기와 폴란드 PO–PiS 갈등·국영기업 후원 네트워크 재구성으로 공식 기록과의 일치를 검증했다.
WikiCLIP은 Wikipedia 기반 오픈 도메인 시각적 엔티티 인식(VER)에서 생성 모델의 높은 계산 비용 문제를 해결하기 위해, LLM 임베딩을 엔티티 표현으로 활용하고 패치 수준의 시각 정보로 이를 정렬하는 Vision-Guided Knowledge Adaptor(VGKA)와 Hard Negative Synthesis 메커니즘을 결합한 대조 학습(contrastive learning) 프레임워크다. OVEN 벤치마크의 unseen 세트에서 기존 최고 성능 대비 16% 향상된 28.5% 정확도를 달성하면서, 생성 모델 AutoVER 대비 추론 지연 시간을 약 100배 단축하는 데 성공했다.
Linking FDA-approved medical devices to their underlying United States Patent and Trademark Office (USPTO) patents enables critical applications such as recall root-cause analysis, M&A-driven IP discovery, and technology trajectory mapping. However, this cross-domain entity linking task remains unexplored due to severe *semantic gaps*: FDA documents focus on clinical outcomes, while patents describe technical mechanisms, yielding minimal lexical overlap. We formalize medical device-patent linking as a challenging cross-domain entity linking problem characterized by label scarcity and domain shifts. Using cardiovascular devices as a high-impact, representative domain featuring diverse technologies, high recall rates, and abundant disclosures, we construct a benchmark with 434 devices, 698K patents, and 585 high-fidelity expert-verified pairs. To address these challenges, we propose Bridge-MedDevKG, a coarse-to-fine framework that integrates (1) **MedDevOnto**, a domain-specific ontology that anchors device concepts via three-tier UMLS normalization; (2) **Multi-signal candidate generation** fusing company affiliation, semantic similarity, and ontology-weighted entity overlap; and (3) **Heterogeneous reranking** with multi-signal scoring and XGBoost classification on hard negatives. Our approach achieves a conservative lower-bound recall of 91.6% on the gold standard with 50.9% noise reduction, substantially outperforming LLM baselines under comparable evaluation. The resulting MedDevKG provides 6.8M high-confidence links, laying a scalable foundation for regulatory-IP integration across medical specialties.
HIPE-2026는 노이즈가 많은 다국어 역사 문헌에서 인물-장소 시간 근거 관계(at, isAt) 추출을 다루는 HIPE 시리즈 3판 공유 과제로, 17개 팀의 40회 이상 제출을 정확도·계산 효율·교차 도메인 일반화 관점에서 분석한다. 프랑스어·독일어·영어 19-20세기 신문과 서프라이즈 문학 도메인에서 OCR 노이즈·역사적 언어 변화·간접 문맥 단서가 핵심 난제이며, 최신 대규모 언어 모델(LLM)부터 경량 분류기까지 다양한 접근의 정확도·효율·견고성 트레이드오프가 드러난다.
정확하고 정보가 풍부한 컬럼 설명(예: 컬럼명 "cust_mem"에 대한 "고객의 멤버십 상태") 생성은 NL2SQL, 테이블 질의응답(table question answering), 엔티티 연결(entity linking) 등 표 형식 데이터에 대한 다양한 하위 자연어 처리(NLP) 작업에 필수적이다. 이 문제는 기업, 도메인 과학, 정부 데이터 포털 등에서 발생한다. 중요성에도 불구하고, 대부분의 실제 데이터셋은 축약된 컬럼명이나 도메인 특화 전문 용어로 인해 문서화가 누락되거나 알기 어려운 경우가 많다. 기존 접근법은 대부분 단일 프롬프트 대규모 언어 모델(LLM)에 의존하며, (i) 약어 처리의 불일치 또는 오류, (ii) 환각(hallucination) 또는 불완전한 설명, (iii) 하위 작업 성능을 저해하는 중복성 또는 모호함의 세 가지 핵심 문제에 어려움을 겪는다. 우리는 LLM을 활용한 자동 컬럼 설명 생성을 위한 작업 인식(task-aware) 프레임워크 TACO를 제안한다. TACO는 3단계 파이프라인을 도입한다: (1) 컬럼명을 표준화하는 약어 확장(abbreviation expansion), (2) 동의어 및 검색 지향 키워드가 풍부한 초기 의미적 설명을 생성하는 설명 생성(description generation), (3) 시뮬레이션된 하위 작업을 활용해 이러한 출력을 정제하는 설명 수정(description revision). 또한, 사람 개입(human-in-the-loop) 확장을 탐구하고 엔티티 연결 및 스키마 보강(schema enrichment)을 위한 새로운 평가 데이터셋을 공개한다. 공개 및 독점 데이터셋에 대한 광범위한 실험 결과, TACO는 기존 방법을 지속적으로 능가하며 하위 작업 성능을 최대 32% 향상시킨다.