• Publications
  • Influence
struc2vec: Learning Node Representations from Structural Identity
TLDR
This work presents struc2vec, a novel and flexible framework for learning latent representations for the structural identity of nodes. Expand
  • 511
  • 81
  • PDF
Ranking Generated Summaries by Correctness: An Interesting but Challenging Application for Natural Language Inference
TLDR
In this paper, we evaluate summaries produced by state-of-the-art models via crowdsourcing and show that such errors occur frequently, in particular with more abstractive models. Expand
  • 49
  • 17
  • PDF
Enhancing AMR-to-Text Generation with Dual Graph Representations
TLDR
We propose a novel graph-to-sequence model that encodes different but complementary perspectives of the structural information contained in the AMR graph. Expand
  • 26
  • 6
  • PDF
Common Sense or World Knowledge? Investigating Adapter-Based Knowledge Injection into Pretrained Transformers
TLDR
We investigate models for complementing the distributional knowledge of BERT with conceptual knowledge from ConceptNet and its corresponding Open Mind Common Sense corpus, respectively, using adapter training. Expand
  • 13
  • 1
  • PDF
Investigating Pretrained Language Models for Graph-to-Text Generation
TLDR
We investigate the impact of large PLMs in graph-to-text generation. Expand
  • 12
  • 1
  • PDF
Modeling Global and Local Node Contexts for Text Generation from Knowledge Graphs
TLDR
We propose novel graph-to-text models that encode an input graph combining both global and local node contexts, in order to learn better contextualized node embeddings. Expand
  • 9
  • 1
  • PDF
Modeling Graph Structure via Relative Position for Better Text Generation from Knowledge Graphs
TLDR
We present a novel encoder-decoder architecture for graph-to-text generation based on Transformer, called the Graformer, which achieves strong performance while using significantly less parameters than other approaches. Expand
  • 6
Análise e Ranqueamento da Rede de Advogados induzida por Processos Judiciais Trabalhistas
Quais sao os advogados mais influentes da justica do trabalho do estado do Rio de Janeiro? Utilizando dados de dezenas de milhares de processos, construimos uma rede de advogados, direcionada e comExpand
  • 1
Modeling Graph Structure via Relative Position for Text Generation from Knowledge Graphs
TLDR
We present Graformer, a novel Transformer-based encoder-decoder architecture for graph-to-text generation, where the encoding of a node relies on all nodes in the input graph - not only direct neighbors. Expand
Structural Adapters in Pretrained Language Models for AMR-to-text Generation
TLDR
We propose STRUCTADAPT, an adapter method to encode graph structure into PLMs, training only 5.1% of the PLM parameters. Expand
...
1
2
...