no code implementations • NAACL (ACL) 2022 • Greta Tuckute, Aalok Sathe, Mingye Wang, Harley Yoder, Cory Shain, Evelina Fedorenko
The modular design of SentSpace allows researchersto easily integrate their own feature computation into the pipeline while benefiting from acommon framework for evaluation and visualization.
1 code implementation • 21 Mar 2024 • Chengxu Zhuang, Evelina Fedorenko, Jacob Andreas
Today's most accurate language models are trained on orders of magnitude more language data than human language learners receive - but with no supervision from other sensory modalities that play a crucial role in human learning.
no code implementations • 21 Mar 2024 • Carina Kauf, Emmanuele Chersoni, Alessandro Lenci, Evelina Fedorenko, Anna A. Ivanova
Experiment 1 shows that, across model architectures and plausibility datasets, (i) log likelihood ($\textit{LL}$) scores are the most reliable indicator of sentence plausibility, with zero-shot prompting yielding inconsistent and typically poor results; (ii) $\textit{LL}$-based performance is still inferior to human performance; (iii) instruction-tuned models have worse $\textit{LL}$-based performance than base models.
no code implementations • 7 Jan 2024 • Greta Tuckute, Dawn Finzi, Eshed Margalit, Joel Zylberberg, SueYeon Chung, Alona Fyshe, Evelina Fedorenko, Nikolaus Kriegeskorte, Jacob Yates, Kalanit Grill Spector, Kohitij Kar
In recent years, neuroscience has made significant progress in building large-scale artificial neural network (ANN) models of brain activity and behavior.
1 code implementation • 28 Nov 2023 • Lukas Wolf, Tiago Pimentel, Evelina Fedorenko, Ryan Cotterell, Alex Warstadt, Ethan Wilcox, Tamar Regev
Using a large spoken corpus of English audiobooks, we extract prosodic features aligned to individual words and test how well they can be predicted from LLM embeddings, compared to non-contextual word embeddings.
no code implementations • 5 Nov 2023 • Eghbal A. Hosseini, Evelina Fedorenko
We quantify straightness using a 1-dimensional curvature metric, and present four findings in support of the trajectory straightening hypothesis: i) In trained models, the curvature decreases from the early to the deeper layers of the network.
no code implementations • 22 Oct 2023 • Jian Li, Greta Tuckute, Evelina Fedorenko, Brian L. Edlow, Adrian V. Dalca, Bruce Fischl
By recognizing the mismatch between geometry and function, JOSA provides new insights into the future development of registration methods using joint analysis of the brain structure and function.
1 code implementation • 20 Oct 2023 • Chengxu Zhuang, Evelina Fedorenko, Jacob Andreas
But to achieve these results, LMs must be trained in distinctly un-human-like ways - requiring orders of magnitude more language data than children receive during development, and without perceptual or social context.
no code implementations • 2 Mar 2023 • Jian Li, Greta Tuckute, Evelina Fedorenko, Brian L. Edlow, Bruce Fischl, Adrian V. Dalca
Brain surface-based image registration, an important component of brain image analysis, establishes spatial correspondence between cortical surfaces.
no code implementations • 16 Jan 2023 • Kyle Mahowald, Anna A. Ivanova, Idan A. Blank, Nancy Kanwisher, Joshua B. Tenenbaum, Evelina Fedorenko
Large Language Models (LLMs) have come closest among all models to date to mastering human language, yet opinions about their linguistic and cognitive capabilities remain split.
1 code implementation • 13 Dec 2022 • Jennifer Hu, Sammy Floyd, Olessia Jouravlev, Evelina Fedorenko, Edward Gibson
We perform a fine-grained comparison of language models and humans on seven pragmatic phenomena, using zero-shot prompting on an expert-curated set of English materials.
1 code implementation • 2 Dec 2022 • Carina Kauf, Anna A. Ivanova, Giulia Rambelli, Emmanuele Chersoni, Jingyuan Selena She, Zawad Chowdhury, Evelina Fedorenko, Alessandro Lenci
Overall, our results show that important aspects of event knowledge naturally emerge from distributional linguistic patterns, but also highlight a gap between representations of possible/impossible and likely/unlikely events.
no code implementations • 23 Aug 2022 • Anna A. Ivanova, Martin Schrimpf, Stefano Anzellotti, Noga Zaslavsky, Evelina Fedorenko, Leyla Isik
Moreover, we argue that, instead of categorically treating the mapping models as linear or nonlinear, we should instead aim to estimate the complexity of these models.
no code implementations • 7 Jun 2022 • Kohitij Kar, Simon Kornblith, Evelina Fedorenko
Given the widespread calls to improve the interpretability of AI systems, we here highlight these different notions of interpretability and argue that the neuroscientific interpretability of ANNs can be pursued in parallel with, but independently from, the ongoing efforts in AI.
no code implementations • 30 Jan 2022 • Kyle Mahowald, Evgeniia Diachek, Edward Gibson, Evelina Fedorenko, Richard Futrell
The conclusion is that grammatical cues such as word order are necessary to convey subjecthood and objecthood in a minority of naturally occurring transitive clauses; nevertheless, they can (a) provide an important source of redundancy and (b) are crucial for conveying intended meaning that cannot be inferred from the words alone, including descriptions of human interactions, where roles are often reversible (e. g., Ray helped Lu/Lu helped Ray), and expressing non-prototypical meanings (e. g., "The bone chewed the dog.
1 code implementation • Proceedings of the National Academy of Sciences 2021 • Martin Schrimpf, Idan Blank, Greta Tuckute, Carina Kauf, Eghbal Hosseini, Nancy Kanwisher, Joshua Tenenbaum, Evelina Fedorenko
The neuroscience of perception has recently been revolutionized with an integrative modeling approach in which computation, brain function, and behavior are linked across many datasets and many computational models.
no code implementations • 29 Sep 2021 • Shashank Srikant, Benjamin Lipkin, Anna A Ivanova, Evelina Fedorenko, Una-May O'Reilly
We find that the Multiple Demand system, a system of brain regions previously shown to respond to code, contains information about multiple specific code properties, as well as machine learned representations of code.
no code implementations • 5 Feb 2018 • Gabriel Grand, Idan Asher Blank, Francisco Pereira, Evelina Fedorenko
Because related words appear in similar contexts, such spaces - called "word embeddings" - can be learned from patterns of lexical co-occurrences in natural language.
1 code implementation • LREC 2018 • Richard Futrell, Edward Gibson, Hal Tily, Idan Blank, Anastasia Vishnevetsky, Steven T. Piantadosi, Evelina Fedorenko
It is now a common practice to compare models of human language processing by predicting participant reactions (such as reading times) to corpora consisting of rich naturalistic linguistic materials.