Building Dynamic Knowledge Graphs From Text Using Machine Reading Comprehension

1. Building Dynamic Knowledge Graphs From Text Using Machine Reading Comprehension Rajarshi Das, Tsendsuren Munkhdalai, Xingdi Yuan, Adam Trischler, Andrew McCallum (ICLR’19) Presented by: Shen Yan

2. Automatically Building Knowledge Graphs • Raw information ⇒ Structured form: • Nodes (entities) • Edges (relationships) • Track the changing relations among entities • Make implicit information more explicit

3. Example:

4. KG-MRC Knowledge Graph - Machine Reading Comprehension • A neural machine-reading model that constructs dynamic knowledge graphs from text • Focus on tracking theLocations of participant entities • Bipartite graph: • Two sets of nodes: entities () and locations ()

5. KG-MRC Pipeline • At each time step , reading prefixes of the paragraph up to and including sentence • Engage Machine reading comprehension (MRC) model to query for the state of each participant entity (e.g., “Where is E located?”). MRC model returns an answer span describing the entities’ current location at , encoding the text span as the vector • Conditioning on the span vector , the model constructs the graph by updating from the previous time step

6. MRC architecture: DrQA(Chen et al. ACL’17）

7. Soft Co-reference • Across time steps: • Within each time step: : incoming location vector : intermediate node representation : matrix of location node representations : matrix of intermediate node representations : co-reference adjacency matrix, to track the related nodes within t

8. Graph Update • Compose all connected nodes with their history summary using an LSTM unit • Update node information • Perform a co-reference pooling operation for location node representations • Recurrentgraph:StackLsuchlayerstopropagatenodeinformationalongthegraph’sedges.

9. Experiments & Evaluation Procedural text comprehension tasks • Task 1 Sentence-level evaluation (Dalvi et al. 2018) • Answer 3 categories of questions • Cat 1: Is E created, (destroyed, ,moved) in the process? • Cat 2: When (step #) is E created, (destroyed, moved)? • Cat 3: Where is E created, (destroyed, moved from/to)? • Task 2 Document-level evaluation (Tandon et al. 2018) • Answer 4 categories of questions • Cat 1: What are the inputs to the process? • Cat 2: What are the outputs of the process? • Cat 3: What conversions occur, when and where? • Cat 4: What movements occur, when and where?

10. Experiments & Evaluation Procedural text comprehension tasks • PROPARA dataset: procedural text about scientific processes.

11. Experiments & Evaluation Procedural text comprehension tasks • PROPARA dataset

12. Experiments & Evaluation 2. Ablation study • Removing the soft-coreference disambiguation within the steps → 1% performance drop • Removing the soft-coreferenceacross time steps → more significant performance drop • ReplacetherecurrentgraphmodulewithLSTM→lacktheinformationpropagationacrossgraphnodes

13. Experiments & Evaluation 3. Commonsense constraints • Commonsense constraints:(Tandonetal.2018) • An entity must exist before it can be moved or destroyed • An entity cannot be created if it already exists • An entity cannot change until it is mentioned in the paragraph

14. Experiments & Evaluation4. Qualitative analysis • Tracking the state of entityblood across 6 sentences • Blue: true location • Orange: predicted results from Pro-Local (Dalvi et al. 2018) • Red: predicted results from KG-MRC

15. Conclusions • Proposedamodelthatconstructsdynamicknowledgegraphsfromtexttotracklocationsofparticipantsentitiesinproceduraltext. • KG-MRCimprovesthedownstreamcomprehensionoftextandachievesstate-of-theartresultsontwoquestion-answeringtasks.

16. Questions?