Giulia Rambelli
grambelli.bsky.social
Giulia Rambelli
@grambelli.bsky.social
🤓 Computational Linguist
🎓 Postdoctoral researcher for @Abstraction_ERC project at @Unibo
🤿 Freediver

https://giuliarambelli.github.io/
Pinned
📣MY FIRST BOOK OUT!📣

Does language rely solely on rules, or is there more to it?
🔎Explore the interplay between compositionality and analogy from a constructionist view in this New Cambridge Element (Elements in #ConstructionGrammar)
👉Download for free: cup.org/41DP95c
⚠️Available just for 4 weeks!
Constructions and Compositionality
Cambridge Core - Grammar and Syntax - Constructions and Compositionality
cup.org
Seeing this slide after completing my postdoc at @abstractionerc.bsky.social makes me emotional. Great talk of @mariannabolog.bsky.social at the workshop "Large and Small Language Models: stato dell’arte fra limiti e prospettive" at Unibo. Many works about abstraction and LLM are on the horizon! :)
November 20, 2025 at 3:21 PM
See you online in 15 minutes :)
📢 This Wed, Nov 19 (10–11:30 AM MT|6PM CET), I’ll be giving a Syntax–Semantics talk at CU Boulder!

Title: Analogy in Language Processing
I’ll discuss how humans + LLMs interpret novel noun–noun compounds, and what this reveals about analogy as a driver of linguistic productivity.

Join us (link ⤵️)📚
November 19, 2025 at 4:39 PM
I'll be here today! ↴
Tomorrow I’m ready to hop on a train to Pisa right after my class, to take part in the round table at CNR ISTI, with top-notch computer scientists, to talk about the use and abuse of AI in research. I’ll be contributing the perspective of the humanities. Stay tuned!
www.isti.cnr.it/it/ricerca/i...
November 18, 2025 at 8:32 AM
📢 This Wed, Nov 19 (10–11:30 AM MT|6PM CET), I’ll be giving a Syntax–Semantics talk at CU Boulder!

Title: Analogy in Language Processing
I’ll discuss how humans + LLMs interpret novel noun–noun compounds, and what this reveals about analogy as a driver of linguistic productivity.

Join us (link ⤵️)📚
November 17, 2025 at 8:02 PM
Unfortunately, I will not attend #EMNLP2025, but you MUST pass by our poster! @tommasoc80.bsky.social will be there explaining our work "Simulating Identity, Propagating Bias: Abstraction and Stereotypes in LLM-Generated Text" with @piasommerauer.bsky.social @emnlpmeeting.bsky.social
With only a week left for #EMNLP2025, we are happy to announce all the works we 🐮 will present 🥳 - come and say "hi" to our posters and presentations during the Main and the co-located events (*SEM and workshops) See you in Suzhou ✈️
November 4, 2025 at 2:55 PM
Reposted by Giulia Rambelli
Evidence that even when LLMs produce similar results to humans, they “rely on lexical associations and statistical priors rather than contextual reasoning or normative criteria. We term this divergence epistemia: the illusion of knowledge emerging when surface plausibility replaces verification”
PNAS
Proceedings of the National Academy of Sciences (PNAS), a peer reviewed journal of the National Academy of Sciences (NAS) - an authoritative source of high-impact, original research that broadly spans...
www.pnas.org
October 17, 2025 at 7:40 AM
Another academic year at @abstractionerc.bsky.social starts with the seminar of prof. Filippo Domaneschi with his talk "Concretely offensive: the processing of reclaimed slurs". Recordings of the talk will be posted here:
site.unibo.it/abstraction/...
October 16, 2025 at 1:47 PM
Reposted by Giulia Rambelli
Filippo Domaneschi, director of the Lab of Language and Cognition presenting at @abstractionerc.bsky.social on the reclamation of slurs, discussing pragmatic and psycholinguistic aspects in relation to abstraction mechanisms
October 16, 2025 at 1:19 PM
✈️ Heading to Düsseldorf for the 2nd #CxGsNLP Workshop at #IWCS2025!
I’ll present a non-archival paper from #ACL2024 + some follow-ups on how LLMs interpret existing and novel compounds.
Excited for the discussions and meeting the vibrant community! On a mission for @abstractionerc.bsky.social
September 23, 2025 at 9:55 AM
Just gave my talk “How Analogy Drives Language Comprehension: The Case of Novel Compounds Interpretation” 🤔🥑🤖

Explored how L1, L2 speakers & LLMs interpret brand-new compounds- L2+LLMs leaning on compositionality, L1 speakers often rely on analogy.

Presented at the #Analogy workshop, #SLE2025
August 29, 2025 at 10:18 AM
Starting now the workshop on Towards a better understanding of #analogy at #SLE2025 !
August 28, 2025 at 10:06 AM
Reposted by Giulia Rambelli
Excited to be at #SLE2025 this week!

Presented 🗣️ two studies:

🔹 Iconicity ratings for Italian language across L1 & L2 speakers

🔹 Ambiguity of generality in a fantastic workshop on "Lexical Ambiguity in the Mind"

Appreciated the feedback & discussions!
August 27, 2025 at 2:02 PM
Abstraction and generalization are 🔥 topics in any linguistic and NLP conference.. and @abstractionerc.bsky.social team is always there to capture the moment! #SLE2025
August 28, 2025 at 8:26 AM
Now attending the PLENARY SESSION at #SLE2025 in Bordeaux 🇫🇷
@dagmardivjak.bsky.social & @petarmilin.bsky.social: Top-down wisdom versus bottom-up noise: can data (ever) replace linguistic expertise?

Excited to hear their insights! #linguistics #conference
August 28, 2025 at 8:13 AM
Here we are at #SLE2025 in Bordeaux! @tomlamarra.bsky.social
August 26, 2025 at 7:26 AM
I will end my postdoc with my first #EMNLP paper! So honored to work with such amazing researchers (and people!)
Could't get a better way to end my summer break: two papers accepted at #EMNLP Findings.
August 25, 2025 at 10:40 AM
✈️ Off to Bordeaux for #SLE2025!
On Friday I’ll present “How Analogy Drives Language Comprehension: The Case of Novel Compounds Interpretation” (a study on humans and LLMs) at the workshop Towards a better understanding of #analogy.

Excited for talks, discussions & some Bordeaux-buzz 🍷
August 25, 2025 at 7:34 AM
Reposted by Giulia Rambelli
Deep Language Learning class 2 now online. Learn about neuroscience, interpretability, and decoding non human communication.

youtu.be/oDpmxmlTofM
Deep Language Learning LSA Summer Institute Class 2
YouTube video by Gasper Begus
youtu.be
July 31, 2025 at 10:21 PM
Reposted by Giulia Rambelli
In a new paper with @giosuebaggio.bsky.social and Eva Wittenberg, we introduce our special issue of Topics in Cognitive Science about "Parallelism in the Architecture of Language" onlinelibrary.wiley.com/doi/10.1111/...
The Present and Future of Parallel Architectures of Language and Cognition
Language as a cognitive capacity involves multiple components, from input and output modalities to formal structures and layers of meaning. This Topic explores the hypothesis that these components fu....
onlinelibrary.wiley.com
August 1, 2025 at 12:29 PM
Reposted by Giulia Rambelli
Flight cancelled?

Thanks to EU action, more online travel agencies will:

✅ Clearly inform you of your rights
✅ Refund flight cancellations within 14 days

This follows a dialogue with the 🇪🇺 and consumer authorities to ensure passengers are protected and EU law, respected.

europa.eu/!gWtYg9
August 1, 2025 at 1:14 PM
Now @melaniemitchell.bsky.social talking about analogy making in humans and AI ! Room 1.34 #ACL2025 @aclmeeting.bsky.social
August 1, 2025 at 2:09 PM
First session is ending, and it is amazing the different approaches, theories, topics proposed by the authors. Have a quick cup of ☕️ and join us for our keynote speaker @KatiaShutova talking about metaphors, politics & humor. #analogy #ACL2025 @piasommerauer.bsky.social @mariannabolog.bsky.social
August 1, 2025 at 8:33 AM
Analogy-angle II just started in room 1.34 at #ACL2025. If you're interest in analogy, pass by! #NLProc @aclmeeting.bsky.social Filip Ilievski @piasommerauer.bsky.social @mariannabolog.bsky.social Ute Schmid
August 1, 2025 at 7:26 AM
About last night at #ACL2025 social dinner! @mariannabolog.bsky.social showed its walzer abilities on the dance floor💃 Always grateful to be part of this community 🙏
July 30, 2025 at 1:09 PM
July 29, 2025 at 7:08 AM