S
Sameer Singh
Researcher at University of California, Irvine
Publications - 196
Citations - 24675
Sameer Singh is an academic researcher from University of California, Irvine. The author has contributed to research in topics: Computer science & Inference. The author has an hindex of 45, co-authored 185 publications receiving 15043 citations. Previous affiliations of Sameer Singh include University of Washington & University of Massachusetts Amherst.
Papers
More filters
Proceedings Article
Explain Your Move: Understanding Agent Actions Using Specific and Relevant Feature Attribution
Nikaash Puri,Sukriti Verma,Piyush Gupta,Dhruv Kayastha,Shripad Deshmukh,Balaji Krishnamurthy,Sameer Singh +6 more
TL;DR: SARFA as mentioned in this paper generates more focused saliency maps by balancing two aspects (specificity and relevance) that capture different desiderata of saliency, i.e., the impact of perturbation on the relative expected reward of the action to be explained.
Posted Content
Universal Adversarial Triggers for NLP.
TL;DR: Universal adversarial triggers as discussed by the authors are input-agnostic sequences of tokens that trigger a model to produce a specific prediction when concatenated to any input from a dataset and are useful for evaluation and interpretation.
Posted Content
Embedding Multimodal Relational Data for Knowledge Base Completion
TL;DR: In this paper, the authors propose multimodal knowledge base embeddings (MKBE) that use different neural encoders for this variety of observed data, and combine them with existing relational models to learn embedding of the entities and multimodi-al data.
Proceedings ArticleDOI
Benefits of Intermediate Annotations in Reading Comprehension
TL;DR: It is observed that for any collection budget, spending a fraction of it on intermediate annotations results in improved model performance, for two complex compositional datasets: DROP and Quoref.
Proceedings Article
Minimally-Supervised Extraction of Entities from Text Advertisements
TL;DR: This paper injects light weight supervision specified as these "constraints" into a semi-Markov conditional random field model of entity extraction in ad creatives using an online learning algorithm and demonstrates significant accuracy improvements on a manually labeled test set as compared to a baseline dictionary approach.