| 1. |
- Beyer, Anne, et al.
(författare)
-
Is Incoherence Surprising? Targeted Evaluation of Coherence Prediction from Language Models
- 2021
-
Ingår i: Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2021, June 6 -11, 2021, Online. - Stroudsburg, PA, USA : Association for Computational Linguistics. - 9781954085466
-
Konferensbidrag (refereegranskat)abstract
- Coherent discourse is distinguished from a mere collection of utterances by the satisfaction of a diverse set of constraints, for example choice of expression, logical relation between denoted events, and implicit compatibility with world-knowledge. Do neural language models encode such constraints? We design an extendable set of test suites addressing different aspects of discourse and dialogue coherence. Unlike most previous coherence evaluation studies, we address specific linguistic devices beyond sentence order perturbations, which allow for a more fine-grained analysis of what constitutes coherence and what neural models trained on a language modelling objective are capable of encoding. Extending the targeted evaluation paradigm for neural language models (Marvin and Linzen, 2018) to phenomena beyond syntax, we show that this paradigm is equally suited to evaluate linguistic qualities that contribute to the notion of coherence.
|
|
| 2. |
- Curzi, Gianluca, 1991, et al.
(författare)
-
A type-assignment of linear erasure and duplication
- 2020
-
Ingår i: Theoretical Computer Science. - 0304-3975.
-
Tidskriftsartikel (refereegranskat)abstract
- We introduce ???, a type-assignment system for the linear λ-calculus that extends second-order ????2, i.e., intuitionistic multiplicative Linear Logic, by means of logical rules that weaken and contract assumptions, but in a purely linear setting. ??? enjoys both a mildly weakened cut-elimination, whose computational cost is cubic, and Subject reduction. A translation of ??? into ????2 exists such that the derivations of the former can exponentially compress the dimension of the derivations in the latter. ??? allows for a modular and compact representation of boolean circuits, directly encoding the fan-out nodes, by contraction, and disposing garbage, by weakening. It can also represent natural numbers with terms very close to standard Church numerals which, moreover, apply to Hereditarily Finite Permutations, i.e. a group structure that exists inside the linear λ-calculus.
|
|
| 3. |
- Curzi, Gianluca, 1991
(författare)
-
Linear Additives
- 2021
-
Ingår i: Electronic Proceedings in Theoretical Computer Science (EPTCS). - 2075-2180.
-
Tidskriftsartikel (refereegranskat)abstract
- We introduce LAM, a subsystem of IMALL2 with restricted additive rules able to manage duplication linearly, called linear additive rules. LAM is presented as the type assignment system for a calculus endowed with copy constructors, which deal with substitution in a linear fashion. As opposed to the standard additive rules, the linear additive rules do not affect the complexity of term reduction: typable terms of LAM enjoy linear strong normalization. Moreover, a mildly weakened version of cut-elimination for this system is proven which takes a cubic number of steps. Finally, we define a sound translation from LAM’s proofs into IMLL2’s linear lambda terms, and we study its complexity.
|
|
| 4. |
- Gren, Lucas, 1984, et al.
(författare)
-
Agility is responsiveness to change: An essential definition
- 2020
-
Ingår i: 24th Conference on Evaluation and Assessment in Software Engineering (EASE2020). - New York, NY, USA : Association for Computing Machinery. - 2153-1633.
-
Konferensbidrag (refereegranskat)abstract
- There is some ambiguity of what agile means in both research and practice. Authors have suggested a diversity of different definitions, through which it is difficult to interpret what agile really is. The concept, however, exists in its implementation through agile practices. In this vision paper, we argue that adopting an agile approach boils down to being more responsive to change. To support this claim, we relate agile principles, practices, the agile manifesto, and our own experiences to this core definition. We envision that agile transformations would be, and are, much easier using this definition and contextualizing its implications.
|
|
| 5. |
- Kalaivaani, P. C. D., et al.
(författare)
-
Advanced lightweight feature interaction in deep neural networks for improving the prediction in click through rate
- 2021
-
Ingår i: Annals of Operations Research. - : Springer. - 0254-5330 .- 1572-9338.
-
Tidskriftsartikel (refereegranskat)abstract
- Online advertising has expanded to a hundred-dollar billion industry in recent years, with sales growing at faster rate in every year. Prediction of the click-through rate (CTR) is an important role in recommended systems and online ads. Click through rating (CTR) is the newest evolution in the advertising and marketing digital world. It is essential for any online advertising company in real time to display the appropriate ads to the right users in the correct context. A huge amount of research work proposed considers each ad separately and does not takes in the relationship with other ads that may have an impact on Click Through Rate. A Factorization machine, a more generalized predictor like support vector machines (SVM) is not able to estimate reliable parameters under sparsity. The main drawback is that the primary features and existing algorithms considers the large weighted parameters. KGCN (Knowledge graph-based convolution network) overcomes the drawback and works on alternating graphs which creates additional clustering and node comparison with high latency and performance. A new framework DeepLight Weight is proposed to resolve the high server latency and high usage of memory issues in online advertising. This work presents a framework to improve the CTR predictions with an objective to accelerate the model inference, prune redundant parameters and the dense embedding vectors. Field Weighed Factorization machine helps to organize the data features with high structure to improve the accuracy. For clearing latency issues, structural pruning makes the algorithm work with dense matrices by combining and executing the individual matrix values or neural nodes.
|
|
| 6. |
- Salahirad, Alireza, et al.
(författare)
-
Mapping the structure and evolution of software testing research over the past three decades
- 2023
-
Ingår i: Journal of Systems and Software. - 0164-1212. ; 195
-
Tidskriftsartikel (refereegranskat)abstract
- Background: The field of software testing is growing and rapidly-evolving. Aims: Based on keywords assigned to publications, we seek to identify predominant research topics and understand how they are connected and have evolved. Methods: We apply co-word analysis to map the topology of testing research as a network where author-assigned keywords are connected by edges indicating co-occurrence in publications. Keywords are clustered based on edge density and frequency of connection. We examine the most popular keywords, summarize clusters into high-level research topics examine how topics connect, and examine how the field is changing. Results: Testing research can be divided into 16 high-level topics and 18 subtopics. Creation guidance, automated test generation, evolution and maintenance, and test oracles have particularly strong connections to other topics, highlighting their multidisciplinary nature. Emerging keywords relate to web and mobile apps, machine learning, energy consumption, automated program repair and test generation, while emerging connections have formed between web apps, test oracles, and machine learning with many topics. Random and requirements-based testing show potential decline. Conclusions: Our observations, advice, and map data offer a deeper understanding of the field and inspiration regarding challenges and connections to explore. Editor's note: Open Science material was validated by the Journal of Systems and Software Open Science Board.
|
|
| 7. |
- Öberg, Johanna
(författare)
-
Deltagandets förutsättningar och digitala resursers påverkan : - ur ett elev- och lärarperspektiv
- 2024
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Denna avhandling undersöker potentialen hos digitala resurser för att stödja elevers deltagande i formella lärmiljöer. Forskningsfrågningar är inriktade på att förstå användarperspektiv, förhållningssätt och krav som är relevanta för att möjliggöra elevers deltagande i skolan. Fyra deltagandebaserade studier undersöker ur ett sociokulturellt perspektiv elevers och lärares syn på teknikförstärkt lärandes roll för att underlätta de betydande interaktioner som främjar deltagande i lärandemiljöer. Elever uttrycker brist på digitala resurser som aktivt stödjer deras deltagande, som till stor del utgår ifrån användandet i deras privatliv. De betonar behovet av ökad användning av digitala resurser i informationshantering, social interaktion och i utbytet med undervisningsinnehåll. Lärare betonar vikten av att kunna ge aktuell och snabb feedback till eleverna via digitala plattformar. Lärarna understryker behovet av att visualisera elevernas resonemang samt kunna sammanställa och ha uppdaterade resultat på både individ- och gruppnivå. Båda lärarna och eleverna identifierar potentialen hos digitala resurser för att öka elevernas deltagande. Det råder enighet om nödvändigheten av att utveckla flexibla digitala resurser som kan hantera informationsflöden och tillhandahålla gemensamma utrymmen för interaktion och samarbete. Resultat från studierna indikerar att lärare och elever har olika uppfattningar om hur mycket digitala resurser används för att stödja elevdeltagandet. Denna slutsats anses grundad i skillnaden mellan elevers och lärares perspektiv på innebörden av elevdeltagande, samt vilka förutsättningar av användningen av digitala resurser som det utgås ifrån (det hemmet erbjuder eller det skolan kan erbjuda). Det sociokulturella perspektivet betonar att lärande underlättas och stärks av ett meningsfullt socialt deltagande, vilket behöver relateras till den pågående utvecklingen av lärmiljöer i det snabbt föränderliga digitala landskapet. Avhandlingen presenterar principer för effektiv användning av teknik med fokus på tillgänglighet, anpassningsförmåga till individuella lärstilar och att stärka en gemensam känsla av bidrag och tillhörighet. Dessa principer, som härrör från lärarnas erfarenheter, tros stödja lärares pedagogiska praktiker i målet att öka elevernas deltagande i lärandeprocessen.
|
|
