AliCG: Fine-grained and Evolvable Conceptual Graph Construction for Semantic Search at Alibaba

Zhang, Ningyu; Jia, Qianghuai; Chen, Xiang; Ye, Hongbin; Chen, Hui; Tou, Huaixiao; Huang, Gang; Wang, Zhao; Hua, Nengwei; Chen, Huajun

doi:10.1145/3447548.3467057

Cited by 34 publications

(10 citation statements)

References 21 publications

(36 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Visual Effect Art Director Film Crew Role Knowledge graphs (KGs) can provide back-end support for a variety of knowledge-intensive tasks in real-world applications, such as recommender systems [17,57], information retrieval [10,50] and time series forecasting [8]. Since KGs usually contain visual information, Multimodal Knowledge Graphs (MKGs) recently have attracted extensive attention in the community of multimedia, natural language processing and knowledge graph [24,38].…”

Section: Superman Returnsmentioning

confidence: 99%

Hybrid Transformer with Multi-level Fusion for Multimodal Knowledge Graph Completion

Chen,

Zhang,

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Multimodal Knowledge Graphs (MKGs), which organize visualtext factual knowledge, have recently been successfully applied to tasks such as information retrieval, question answering, and recommendation system. Since most MKGs are far from complete, extensive knowledge graph completion studies have been proposed focusing on the multimodal entity, relation extraction and link prediction. However, different tasks and modalities require changes to the model architecture, and not all images/objects are relevant to text input, which hinders the applicability to diverse real-world scenarios. In this paper, we propose a hybrid transformer with multi-level fusion to address those issues. Specifically, we leverage a hybrid transformer architecture with unified input-output for diverse multimodal knowledge graph completion tasks. Moreover, we propose multi-level fusion, which integrates visual and text representation via coarse-grained prefix-guided interaction and fine-grained correlation-aware fusion modules. We conduct extensive experiments to validate that our MKGformer can obtain SOTA performance on four datasets of multimodal link prediction, multimodal RE, and multimodal NER 1 .

show abstract

Section: Superman Returnsmentioning

confidence: 99%

Hybrid Transformer with Multi-level Fusion for Multimodal Knowledge Graph Completion

Chen,

Zhang,

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…PLMs recently have significant impact on NER (Zhang et al 2021), where Transformer-based models (Peters et al 2018;Devlin et al 2019;Zheng et al 2021;Nan et al 2021) are utilized as backbone network for acquiring plentiful representations. The current dominant methods (Chiu and Nichols 2016;Ma and Hovy 2016;Liu et al 2019;Strubell et al 2017;Zhang et al 2020a;Liu et al 2021a,b) treat NER as a sequence tagging problem with label-specific classifiers or CRF.…”

Section: Nermentioning

confidence: 99%

LightNER: A Lightweight Tuning Paradigm for Low-resource NER via Pluggable Prompting

Chen¹,

Li²,

Deng³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Most existing NER methods rely on extensive labeled data for model training, which struggles in the low-resource scenarios with limited training data. Recently, prompt-tuning methods for pre-trained language models have achieved remarkable performance in few-shot learning by exploiting prompts as task guidance to reduce the gap between training progress and downstream tuning. Inspired by prompt learning, we propose a novel lightweight generative framework with promptguided attention for low-resource NER (LightNER). Specifically, we construct the semantic-aware answer space of entity categories for prompt learning to generate the entity span sequence and entity categories without any label-specific classifiers. We further propose prompt-guided attention by incorporating continuous prompts into the self-attention layer to re-modulate the attention and adapt pre-trained weights. Note that we only tune those continuous prompts with the whole parameter of the pre-trained language model fixed, thus, making our approach lightweight and flexible for low-resource scenarios and can better transfer knowledge across domains. Experimental results show that LightNER can obtain comparable performance in the standard supervised setting and outperform strong baselines in low-resource settings by tuning only a small part of the parameters.

show abstract

“…Rule2: Function Mutual Exclusion. Given the generated logical form L, with the entire function set O, and the default 1 We have experiment with Logic2T ext with automatically produced logical forms, but obtain little improvement.…”

Section: Self-trainingmentioning

confidence: 99%

“…Natural language generation (NLG) from structured data has good application prospects in communicating with humans in a natural way [1], such as financial report [2], medical report [3] and so on. However, previous studies [4] mostly concentrate on surface descriptions from simple records, such as limited schema (e.g., E2E [5], and WikiBio [6]), which suffer from low fidelity and uncontrollable content selection.…”

Section: Introductionmentioning

confidence: 99%

LOGEN: Few-shot Logical Knowledge-Conditioned Text Generation with Self-training

Zhang¹,

Ye²,

Yang³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Natural language generation from structured data mainly focuses on surface-level descriptions, suffering from uncontrollable content selection and low fidelity. Previous works leverage logical forms to facilitate logical knowledge-conditioned text generation. Though achieving remarkable progress, they are data-hungry, which makes the adoption for real-world applications challenging with limited data. To this end, this paper proposes a unified framework for logical knowledge-conditioned text generation in the few-shot setting. With only a few seeds logical forms (e.g., 20/100 shot), our approach leverages selftraining and samples pseudo logical forms based on content and structure consistency. Experimental results demonstrate that our approach can obtain better few-shot performance than baselines.

show abstract

AliCG: Fine-grained and Evolvable Conceptual Graph Construction for Semantic Search at Alibaba

Cited by 34 publications

References 21 publications

Hybrid Transformer with Multi-level Fusion for Multimodal Knowledge Graph Completion

Hybrid Transformer with Multi-level Fusion for Multimodal Knowledge Graph Completion

LightNER: A Lightweight Tuning Paradigm for Low-resource NER via Pluggable Prompting

LOGEN: Few-shot Logical Knowledge-Conditioned Text Generation with Self-training

Contact Info

Product

Resources

About