Results 1 to 10 of about 10,598,445 (199)
Many Task Learning with Task Routing [PDF]
2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019 Typical multi-task learning (MTL) methods rely on architectural adjustments and a large trainable parameter set to jointly optimize over several tasks. However, when the number of tasks increases so do the complexity of the architectural adjustments and ...
Strezoski, Gjorgji +2 more
core +3 more sources
Editing Models with Task Arithmetic [PDF]
International Conference on Learning Representations, 2022 Changing how pre-trained models behave -- e.g., improving their performance on a downstream task or mitigating biases learned during pre-training -- is a common practice when developing machine learning systems.
Gabriel Ilharco +6 more
semanticscholar +1 more source
GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding [PDF]
BlackboxNLP@EMNLP, 2018 Human ability to understand language is general, flexible, and robust. In contrast, most NLU models above the word level are designed for a specific task and struggle with out-of-domain data.
Alex Wang +5 more
semanticscholar +1 more source
Multi-task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics [PDF]
2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2017 Numerous deep learning applications benefit from multitask learning with multiple regression and classification objectives. In this paper we make the observation that the performance of such systems is strongly dependent on the relative weighting between
Alex Kendall, Y. Gal, R. Cipolla
semanticscholar +1 more source
Multiagent Task Coordination as Task Allocation Plus Task Responsibility [PDF]
, 2020 In this work, we present a dynamic Task Coordination framework () for multiagent systems. Here task coordination refers to a twofold problem where an exogenously imposed state of affairs should be satisfied by a multiagent system. To address this problem the involved agents or agent groups need to be assigned tasks to fulfill (task allocation) and the ...
Vahid Yazdanpanah +5 more
openaire +2 more sources
ProgPrompt: Generating Situated Robot Task Plans using Large Language Models [PDF]
IEEE International Conference on Robotics and Automation, 2022 Task planning can require defining myriad domain knowledge about the world in which a robot needs to act. To ameliorate that effort, large language models (LLMs) can be used to score potential next actions during task planning, and even generate action ...
Ishika Singh +8 more
semanticscholar +1 more source
Multi-Task Retrieval for Knowledge-Intensive Tasks [PDF]
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), 2021 Retrieving relevant contexts from a large corpus is a crucial step for tasks such as open-domain question answering and fact checking. Although neural retrieval outperforms traditional methods like tf-idf and BM25, its performance degrades considerably when applied to out-of-domain data. Driven by the question of whether a neural retrieval model can be
Jean Maillard +6 more
openaire +2 more sources
TOOD: Task-aligned One-stage Object Detection [PDF]
IEEE International Conference on Computer Vision, 2021 One-stage object detection is commonly implemented by optimizing two sub-tasks: object classification and localization, using heads with two parallel branches, which might lead to a certain level of spatial misalignment in predictions between the two ...
Chengjian Feng +4 more
semanticscholar +1 more source
Task inhibition and task repetition in task switching [PDF]
European Journal of Cognitive Psychology, 2006 In task-switching experiments with three tasks, the relative cost of an N–2 task repetition (task sequence ABA) compared to a task switch (task sequence CBA) is referred to as N–2 repetition cost. N–2 repetition cost is assumed to reflect persisting inhibition of a task that was recently switched away from.
Philipp, A., Koch, I.
openaire +2 more sources
Introduction to the CoNLL-2003 Shared Task: Language-Independent Named Entity Recognition [PDF]
Conference on Computational Natural Language Learning, 2003 We describe the CoNLL-2003 shared task: language-independent named entity recognition. We give background information on the data sets (English and German) and the evaluation method, present a general overview of the systems that have taken part in the ...
E. Tjong Kim Sang, F. D. Meulder
semanticscholar +1 more source