29 citations · 53 across the 18 of their papers we have counts for
20 papers
Knowledge Offloading: Decomposing LLMs into Sparse Backbones and Memory Modules
Karim Galliamov, Rochelle Choenni, Ivan Titov
LLMs encode both general capabilities and domain-specific knowledge in a single set of parameters. We ask whether this capacity can be reorganized: keeping broadly useful computati…
Finding Culture-Sensitive Neurons in Vision-Language Models
Xiutian Zhao, Rochelle Choenni, Rohit Saxena +1
Despite their impressive performance, vision-language models (VLMs) still struggle on culturally situated inputs. To understand how VLMs process culturally grounded information, we…
Best-of-L: Cross-Lingual Reward Modeling for Mathematical Reasoning
Sara Rajaee, Rochelle Choenni, Ekaterina Shutova +1
While the reasoning abilities of large language models (LLMs) continue to advance, it remains unclear how such ability varies across languages in multilingual LLMs and whether diff…
M-Wanda: Improving One-Shot Pruning for Multilingual LLMs
Rochelle Choenni, Ivan Titov
Multilingual LLM performance is often critically dependent on model size. With an eye on efficiency, this has led to a surge in interest in one-shot pruning methods that retain the…
Local Contrastive Editing of Gender Stereotypes
Marlene Lutz, Rochelle Choenni, Markus Strohmaier +1
Stereotypical bias encoded in language models (LMs) poses a threat to safe language technology, yet our understanding of how bias manifests in the parameters of LMs remains incompl…
Self-Alignment: Improving Alignment of Cultural Values in LLMs via In-Context Learning
Rochelle Choenni, Ekaterina Shutova
Improving the alignment of Large Language Models (LLMs) with respect to the cultural values that they encode has become an increasingly important topic. In this work, we study whet…
On the Evaluation Practices in Multilingual NLP: Can Machine Translation Offer an Alternative to Human Translations?
Rochelle Choenni, Sara Rajaee, Christof Monz +1
While multilingual language models (MLMs) have been trained on 100+ languages, they are typically only evaluated across a handful of them due to a lack of available test data in mo…
The Echoes of Multilinguality: Tracing Cultural Value Shifts during LM Fine-tuning
Rochelle Choenni, Anne Lauscher, Ekaterina Shutova
Texts written in different languages reflect different culturally-dependent beliefs of their writers. Thus, we expect multilingual LMs (MLMs), that are jointly trained on a concate…
Metaphor Understanding Challenge Dataset for LLMs
Xiaoyu Tong, Rochelle Choenni, Martha Lewis +1
Metaphors in natural language are a reflection of fundamental cognitive processes such as analogical reasoning and categorisation, and are deeply rooted in everyday communication.…
Examining Modularity in Multilingual LMs via Language-Specialized Subnetworks
Rochelle Choenni, Ekaterina Shutova, Dan Garrette
Recent work has proposed explicitly inducing language-wise modularity in multilingual LMs via sparse fine-tuning (SFT) on per-language subnetworks as a means of better guiding cros…
Do large language models solve verbal analogies like children do?
Claire E. Stevenson, Mathilde ter Veen, Rochelle Choenni +2
Analogy-making lies at the heart of human cognition. Adults solve analogies such as \textit{Horse belongs to stable like chicken belongs to ...?} by mapping relations (\textit{kept…
Probing LLMs for Joint Encoding of Linguistic Categories
Giulio Starace, Konstantinos Papakostas, Rochelle Choenni +4
Large Language Models (LLMs) exhibit impressive performance on a range of NLP tasks, due to the general-purpose linguistic knowledge acquired during pretraining. Existing model int…
How do languages influence each other? Studying cross-lingual data sharing during LM fine-tuning
Rochelle Choenni, Dan Garrette, Ekaterina Shutova
Multilingual large language models (MLLMs) are jointly trained on data from many different languages such that representation of individual languages can benefit from other languag…
Data-Efficient Cross-Lingual Transfer with Language-Specific Subnetworks
Rochelle Choenni, Dan Garrette, Ekaterina Shutova
Large multilingual language models typically share their parameters across all languages, which enables cross-lingual task transfer, but learning can also be hindered when training…
Stepmothers are mean and academics are pretentious: What do pretrained language models learn about you?
Rochelle Choenni, Ekaterina Shutova, Robert van Rooij
In this paper, we investigate what types of stereotypical information are captured by pretrained language models. We present the first dataset comprising stereotypical attributes o…
Cross-neutralising: Probing for joint encoding of linguistic information in multilingual models
Rochelle Choenni, Ekaterina Shutova
Multilingual sentence encoders are widely used to transfer NLP models across languages. The success of this transfer is, however, dependent on the model's ability to encode the pat…
What does it mean to be language-agnostic? Probing multilingual sentence encoders for typological properties
Rochelle Choenni, Ekaterina Shutova
Multilingual sentence encoders have seen much success in cross-lingual model transfer for downstream NLP tasks. Yet, we know relatively little about the properties of individual la…
Blackbox meets blackbox: Representational Similarity and Stability Analysis of Neural Language Models and Brains
Samira Abnar, Lisa Beinborn, Rochelle Choenni +1
In this paper, we define and apply representational stability analysis (ReStA), an intuitive way of analyzing neural language models. ReStA is a variant of the popular representati…
Robust Evaluation of Language-Brain Encoding Experiments
Lisa Beinborn, Samira Abnar, Rochelle Choenni
Language-brain encoding experiments evaluate the ability of language models to predict brain responses elicited by language stimuli. The evaluation scenarios for this task have not…
Semantic Drift in Multilingual Representations
Lisa Beinborn, Rochelle Choenni
Multilingual representations have mostly been evaluated based on their performance on specific tasks. In this article, we look beyond engineering goals and analyze the relations be…