More Like this

1. Dialog Policy Learning for Joint Clarification and Active Learning Queries

   Padmakumar, A; Mooney, R. (February 2021, Proceedings of the AAAI Conference on Artificial Intelligence)

   null (Ed.)

   Intelligent systems need to be able to recover from mistakes, resolve uncertainty, and adapt to novel concepts not seen during training. Dialog interaction can enable this by the use of clarifications for correction and resolving uncertainty, and active learning queries to learn new concepts encountered during operation. Prior work on dialog systems has either focused on exclusively learning how to perform clarification/ information seeking, or to perform active learning. In this work, we train a hierarchical dialog policy to jointly perform both clarification and active learning in the context of an interactive language-based image retrieval task motivated by an on-line shopping application, and demonstrate that jointly learning dialog policies for clarification and active learning is more effective than the use of static dialog policies for one or both of these functions. 

   more » « less
2. Opportunistic Active Learning for Grounding Natural Language Descriptions

   Thomason, J.; Padmakumar, A.; Sinapov, J.; Hart, J.; Stone, P.; Mooney, R.J. (November 2017, Proceedings of the 1st Annual Conference on Robot Learning (CoRL-17))

   Active learning identifies data points from a pool of unlabeled examples whose labels, if made available, are most likely to improve the predictions of a supervised model. Most research on active learning assumes that an agent has access to the entire pool of unlabeled data and can ask for labels of any data points during an initial training phase. However, when incorporated in a larger task, an agent may only be able to query some subset of the unlabeled pool. An agent can also opportunistically query for labels that may be useful in the future, even if they are not immediately relevant. In this paper, we demonstrate that this type of opportunistic active learning can improve performance in grounding natural language descriptions of everyday objects---an important skill for home and office robots. We find, with a real robot in an object identification setting, that inquisitive behavior---asking users important questions about the meanings of words that may be off-topic for the current dialog---leads to identifying the correct object more often over time. 

   more » « less
3. End-to-end Knowledge Retrieval with Multi-modal Queries

   Luo, Man; Fang, Zhiyuan; Gokhale, Tejas; Yang, Yezhou; Baral, Chitta (July 2023, 61st Annual Meeting of the Association for Computational Linguistics)

   We investigate knowledge retrieval with multi-modal queries, i.e. queries containing information split across image and text inputs, a challenging task that differs from previous work on cross-modal retrieval. We curate a new dataset called ReMuQ for benchmarking progress on this task. ReMuQ requires a system to retrieve knowledge from a large corpus by integrating contents from both text and image queries. We introduce a retriever model “ReViz” that can directly process input text and images to retrieve relevant knowledge in an end-to-end fashion without being dependent on intermediate modules such as object detectors or caption generators. We introduce a new pretraining task that is effective for learning knowledge retrieval with multimodal queries and also improves performance on downstream tasks. We demonstrate superior performance in retrieval on two datasets (ReMuQ and OK-VQA) under zero-shot settings as well as further improvements when finetuned on these datasets. 

   more » « less
4. CoaCor: Code Annotation for Code Retrieval with Reinforcement Learning

   https://doi.org/10.1145/3308558.3313632  

   Yao, Ziyu; Peddamail, Jayavardhan Reddy; Sun, Huan (January 2019, The World Wide Web Conference)

   To accelerate software development, much research has been performed to help people understand and reuse the huge amount of available code resources. Two important tasks have been widely studied: code retrieval, which aims to retrieve code snippets relevant to a given natural language query from a code base, and code annotation, where the goal is to annotate a code snippet with a natural language description. Despite their advancement in recent years, the two tasks are mostly explored separately. In this work, we investigate a novel perspective of Code annotation for Code retrieval (hence called “CoaCor”), where a code annotation model is trained to generate a natural language annotation that can represent the semantic meaning of a given code snippet and can be leveraged by a code retrieval model to better distinguish relevant code snippets from others. To this end, we propose an effective framework based on reinforcement learning, which explicitly encourages the code annotation model to generate annotations that can be used for the retrieval task. Through extensive experiments, we show that code annotations generated by our framework are much more detailed and more useful for code retrieval, and they can further improve the performance of existing code retrieval models significantly. 

   more » « less
5. Anderson and Krathwohl's Two-Dimensional Taxonomy Applied to Task Creation and Learning Assessment

   https://doi.org/10.1145/3341981.3344226  

   Urgo, K.; Arguello, J.; Capra, R. (October 2019, Proceedings of the 5th ACM SIGIR International Conference on the Theory of Information Retrieval)

   Search tasks play an important role in the study and development of interactive information retrieval (IIR) systems. Prior work has examined how search tasks vary along dimensions such as the task’s main activity, end goal, structure, and complexity. Recently, researchers have been exploring task complexity from the perspective of cognitive complexity—related to the types (and variety) of mental activities required by the task. Anderson & Krathwohl’s two-dimensional taxonomy of learning has been a commonly used framework for investigating tasks from the perspective of cognitive complexity [1]. A&K’s 2D taxonomy involves a cognitive process dimension and an orthogonal knowledge dimension. Prior IIR research has successfully leveraged the cognitive process dimension of this 2D taxonomy to develop search tasks and investigate their effects on searchers’ needs, perceptions, and behaviors. However, the knowledge dimension of the taxonomy has been largely ignored. In this conceptual paper, we argue that future IIR research should consider both dimensions of A&K’s taxonomy. Specifically, we discuss related work, present details on both dimensions of A&K’s taxonomy, and explain how to use the taxonomy to develop search tasks and learning assessment materials. Additionally, we discuss how considering both dimensions of A&K’s taxonomy has important implications for future IIR research. 

   more » « less