Brain Score remains similar when language models are trained on diverse natural languages or on structured non-language data like DNA and code, indicating the metric tracks shared structural extraction but is not diagnostic of human-like language processing.
Najoung Kim, Sebastian Schuster, and Shubham Toshniwal
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
verdicts
UNVERDICTED 5representative citing papers
Introduces natural identifiers (NIDs) from common training data to support post-hoc differential privacy auditing and dataset inference for LLMs without retraining or private held-out sets.
Causal interventions reveal a retrieval-conditioned rebinding circuit in attention heads that supports dynamic entity tracking in Gemma and Llama models, with family-specific representational signatures.
RExBench is a new benchmark showing that LLM coding agents fail to autonomously implement most realistic research extensions to prior AI papers.
Language models can support formal generative linguistic theories, expanding testable theories and potentially reconciling them with usage-based accounts.
citing papers explorer
-
Brain Score Tracks Shared Properties of Languages: Evidence from Many Natural Languages and Structured Sequences
Brain Score remains similar when language models are trained on diverse natural languages or on structured non-language data like DNA and code, indicating the metric tracks shared structural extraction but is not diagnostic of human-like language processing.
-
Natural Identifiers for Privacy and Data Audits in Large Language Models
Introduces natural identifiers (NIDs) from common training data to support post-hoc differential privacy auditing and dataset inference for LLMs without retraining or private held-out sets.
-
A retrieval conditioned rebinding circuit for dynamic entity tracking in large language models
Causal interventions reveal a retrieval-conditioned rebinding circuit in attention heads that supports dynamic entity tracking in Gemma and Llama models, with family-specific representational signatures.
-
RExBench: Can coding agents autonomously implement AI research extensions?
RExBench is a new benchmark showing that LLM coding agents fail to autonomously implement most realistic research extensions to prior AI papers.
-
Not-So-Strange Love: Language Models and Generative Linguistic Theories are More Compatible than They Appear
Language models can support formal generative linguistic theories, expanding testable theories and potentially reconciling them with usage-based accounts.