The conditional directed acyclic graph (DAG) task model can represent the conditional execution flows that commonly exist in many real-time parallel applications. Previous work has shown that by properly assigning the priority among vertices inside a nonconditional DAG task, we can reduce the task response time and achieve better system schedulability. This article studies how to apply intra-task priority assignment to conditional DAG tasks. We develop a response time bound that theoretically dominates the state-of-the-art and present a novel algorithm to compute the bound in polynomial time. We further extend the proposed approach to the general setting of multiple conditional DAG tasks. Experiments with one conditional DAG task and multiple conditional DAG tasks demonstrate that our method consistently outperforms the state-of-the-art by a considerable margin.

Directed acyclic graph (DAG) becomes a popular model for modern real-time embedded software. It is really a challenge to bound the worst-case response time (WCRT) of DAG task. Parallelism, dependencies and mutual exclusion become three of the most critical properties of real-time parallel tasks. Recent work applied prioritizing techniques to reduce DAG task's WCRT bound, which has well studied the first two properties, i.e., parallelism and dependencies, but leaves the mutually exclusive property as an open problem. This paper focuses on all the three properties of real-time parallel software, and investigates how to estimate the WCRT of the DAG task model with mutually exclusive vertices and under prioritized list scheduling algorithms. We derive a reasonable WCRT bound for such a complicated DAG task, and prove that the corresponding WCRT bound computation problem is strongly NP-hard. It means that there are no pseudo-polynomial time algorithms to compute the WCRT bound. For the prioritized DAG with a constant number of mutual exclusive vertices, we develop a dynamic programming algorithm that is able to estimate the WCRT bound within pseudo-polynomial time. Experiments are conducted to evaluate the performance of our analysis method implemented with different priority assignment policies against the state-of-the-art.

Span extraction, aiming to extract text spans (such as words or phrases) from plain texts, is a fundamental process in Information Extraction.Recent works introduce the label knowledge to enhance the text representation by formalizing the span extraction task into a question answering problem (QA Formalization), which achieves state-of-the-art performance.However, QA Formalization does not fully exploit the label knowledge and suffers from low efficiency in training/inference.To address those problems, we introduce a new paradigm to integrate label knowledge and further propose a novel model to explicitly and efficiently integrate label knowledge into text representations.Specifically, it encodes texts and label annotations independently and then integrates label knowledge into text representation with an elaborate-designed semantics fusion module.We conduct extensive experiments on three typical span extraction tasks: flat NER, nested NER, and event detection.The empirical results show that 1) our method achieves state-of-the-art performance on four benchmarks, and 2) reduces training time and inference time by 76% and 77% on average, respectively, compared with the QA Formalization paradigm.Our code and data are available at https://github.com/ Akeepers/LEAR.

Even though earliest-deadline-first (EDF) is optimal in terms of uniprocessor schedulability, it is co-NP-hard to precisely verify uniprocessor schedulability for constrained-deadline task sets. The most efficient way to solve this problem in polynomial time is via a partially linear approximation of the demand bound function. Such approximation leads to a simple uniprocessor schedulability testing with speedup factor ρ. Such a result further leads to Deadline-Monotonic Partitioned-EDF on multi-processors with speedup factor of 1 + ρ − 1/m (where m is the number of processors). The current state of the art results indicate that ρ is within the range [1.5,14/9]. Especially, it has been a conjecture that ρ = 1.5.This paper improves the range of ρ to (1.5026,1.5380). The improved lower bound disproves the conjecture of lower bound 1.5. A novel technique is to construct an auxiliary function that is larger than the approximate demand bound function but keeps the supremum ρ unchanged. It solves the dilemma that beating the lower bound 1.5 requires extremely large task sets, while the large size makes it difficult to check the schedulability. This technique not only enables us to disprove 1.5 by a task set of only eight tasks, but also sheds light on future work in transferring/downsizing task sets and deriving utilization bound based tests for various workload abstraction models, such as DAG tasks.