• Publications
  • Influence
On Generating Characteristic-rich Question Sets for QA Evaluation
tl;dr
We present a semi-automated framework for constructing factoid question answering (QA) datasets, where an array of question characteristics are formalized, including structure complexity, function, commonness, answer cardinality, and paraphrasing. Expand
  • 47
  • 9
  • Open Access
Building Natural Language Interfaces to Web APIs
tl;dr
We study the problem of natural language interface to APIs (NL2APIs), with a focus on web APIs for web services. Expand
  • 22
  • 5
  • Open Access
Table Cell Search for Question Answering
tl;dr
We investigate an important yet largely under-addressed problem: Given millions of tables, how to precisely retrieve table cells to answer a user question. Expand
  • 51
  • 4
  • Open Access
Cross-domain Semantic Parsing via Paraphrasing
tl;dr
We formulate cross-domain semantic parsing as a domain adaptation problem: train a semantic parser on some source domains and then adapt it to the target domain. Expand
  • 46
  • 2
  • Open Access
DialSQL: Dialogue Based Structured Query Generation
tl;dr
We introduce DialSQL, a dialoguebased structured query generation framework that leverages human intelligence to boost the performance of existing algorithms via user interaction. Expand
  • 29
  • 2
  • Open Access
Improving Semantic Parsing via Answer Type Inference
tl;dr
In this work, we show the possibility of inferring the answer type before solving a factoid question and leveraging the type information to improve semantic parsing. Expand
  • 28
  • 2
  • Open Access
Learning to Compose Topic-Aware Mixture of Experts for Zero-Shot Video Captioning
tl;dr
We propose a principled Topic-Aware Mixture of Experts (TAMoE) model for zero-shot video captioning, which learns to compose different experts based on different topic embeddings, implicitly transferring the knowledge learned from seen activities to unseen ones. Expand
  • 17
  • 2
  • Open Access
An End-to-End Deep Framework for Answer Triggering with a Novel Group-Level Objective
tl;dr
We propose an end-to-end deep neural network framework, which is trained by a novel group-level objective function that directly optimizes the answer triggering performance. Expand
  • 8
  • 2
  • Open Access
On the validity of geosocial mobility traces
tl;dr
A large portion of visited locations is missing from Foursquare checkins, and most checkin events are either forged or superfluous events. Expand
  • 40
  • 1
  • Open Access
Global Relation Embedding for Relation Extraction
tl;dr
We study the problem of textual relation embedding with distant supervision. Expand
  • 16
  • 1
  • Open Access