Welcome to the 2nd edition of the Workshop on Analogical Abstraction in Cognition, Perception, and Language (Analogy-Angle II)! Analogical abstraction is a fundamental human cognitive skill (Penn et al. 2008) (Penn et al. 2008; Hofstadter 2001) defined as the ability to perceive and utilize the similarities between concepts, situations, or events based on (systems of) relations rather than surface similarities (Holyoak 2012; Gentner and Smith 2012). Analogy enables creative inferences, explanations, and generalization of knowledge and has been used for scientific inventions (Dunbar 2012), solving problems (Gick and Holyoak 1980), and policy-making Houghton 1998. As such, it has been the subject of cognitive theories and studies about humans for standard processes, such as retrieving memories (Wharton et al. 1994) and problem solving (Gick and Holyoak 1980). Analogical tasks have gained considerable popularity in natural language processing (NLP) and artificial intelligence (AI), where they are often framed as tests of a model’s intelligence in comparison to human performance. These tasks typically involve so-called word-based proportional analogies of the form (A : B :: C : D).(Mikolov et al. 2013a; https://arxiv.org/pdf/1301.3781; Gladkova et al. 2016; Ushio et al. 2021) lend themselves well to large language models (LLMs) (Webb et al. 2023). However, controlling for association and memorization (Stevenson et al. 2023; Lewis and Mitchell 2024) or shifting toward more complex settings like narratives reveals limitations in scope, generalizability, and alignment with cognitive theories (Nagarajah et al. 2022; Wijesiriwardene et al. 2023; Sourati et al. 2024). Inspired by the richness of analogical abstraction and the wide interest in this topic from computational linguistics, artificial intelligence, and cognitive psychology, ANALOGY-ANGLE II connects these communities and facilitates cross-disciplinary activities. ANALOGY-ANGLE II welcomes novel contributions in short, long, and review formats, as well as relevant papers accepted at top-tier venues over the past year (so-called dissemination papers).
Rambelli, G., Ilievski, F., Bolognesi, M., Sommerauer, P. (2025). Proceedings of the 2nd Workshop on Analogical Abstraction in Cognition, Perception, and Language (Analogy-Angle II). Kerrville, TX : Association for Computational Linguistics [10.18653/v1/2025.analogyangle-1].
Proceedings of the 2nd Workshop on Analogical Abstraction in Cognition, Perception, and Language (Analogy-Angle II)
Giulia RambelliPrimo
Membro del Collaboration Group
;Marianna BolognesiMembro del Collaboration Group
;
2025
Abstract
Welcome to the 2nd edition of the Workshop on Analogical Abstraction in Cognition, Perception, and Language (Analogy-Angle II)! Analogical abstraction is a fundamental human cognitive skill (Penn et al. 2008) (Penn et al. 2008; Hofstadter 2001) defined as the ability to perceive and utilize the similarities between concepts, situations, or events based on (systems of) relations rather than surface similarities (Holyoak 2012; Gentner and Smith 2012). Analogy enables creative inferences, explanations, and generalization of knowledge and has been used for scientific inventions (Dunbar 2012), solving problems (Gick and Holyoak 1980), and policy-making Houghton 1998. As such, it has been the subject of cognitive theories and studies about humans for standard processes, such as retrieving memories (Wharton et al. 1994) and problem solving (Gick and Holyoak 1980). Analogical tasks have gained considerable popularity in natural language processing (NLP) and artificial intelligence (AI), where they are often framed as tests of a model’s intelligence in comparison to human performance. These tasks typically involve so-called word-based proportional analogies of the form (A : B :: C : D).(Mikolov et al. 2013a; https://arxiv.org/pdf/1301.3781; Gladkova et al. 2016; Ushio et al. 2021) lend themselves well to large language models (LLMs) (Webb et al. 2023). However, controlling for association and memorization (Stevenson et al. 2023; Lewis and Mitchell 2024) or shifting toward more complex settings like narratives reveals limitations in scope, generalizability, and alignment with cognitive theories (Nagarajah et al. 2022; Wijesiriwardene et al. 2023; Sourati et al. 2024). Inspired by the richness of analogical abstraction and the wide interest in this topic from computational linguistics, artificial intelligence, and cognitive psychology, ANALOGY-ANGLE II connects these communities and facilitates cross-disciplinary activities. ANALOGY-ANGLE II welcomes novel contributions in short, long, and review formats, as well as relevant papers accepted at top-tier venues over the past year (so-called dissemination papers).| File | Dimensione | Formato | |
|---|---|---|---|
|
unpaywall-bitstream--1266994566.pdf
accesso aperto
Tipo:
Versione (PDF) editoriale / Version Of Record
Licenza:
Creative commons
Dimensione
9.18 MB
Formato
Adobe PDF
|
9.18 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.



