Papers by Jannes Münchmeyer
Extend, don’t rebuild: Phrasing conditional graph modification as autoregressive sequence labelling (2021.emnlp-main)
Copied to clipboard
| Challenge: | Generating or modifying graphs from natural language text has applications in many subfields, such as dependency parsing or knowledge graph construction. |
| Approach: | They propose a method that first embeds the graph and the instructions with a joint encoder and then rebuilds it using a separate generative model for graphs conditioned on h. |
| Outcome: | The proposed method improves accuracy on three scene graph modification data sets while the state-of-the-art fails to generalize. |
NLProlog: Reasoning with Weak Unification for Question Answering in Natural Language (P19-1)
Copied to clipboard
| Challenge: | ambiguity in natural language is difficult to interpret due to large linguistic variability. |
| Approach: | They propose to use a Prolog prover to extend neural networks with logic programming to solve multi-hop reasoning tasks over natural language. |
| Outcome: | The proposed model outperforms baseline models on two question answering tasks and is competitive on the MedHop corpus. |