Lesioning parameters in large language models produces aphasia-like symptoms whose distributions vary by attention versus feed-forward components and by layer depth, but differ qualitatively from human clinical profiles.
Title resolution pending
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4representative citing papers
Larger LLMs hallucinate more often despite having the correct concept available because instruction tuning causes probability mass to disperse across alternative surface forms instead of concentrating on one.
The paper introduces the Construct Validity Protocol to validate semantic embeddings for social constructs and proposes Counterfactual Neutralization using LLMs to reduce confounding.
citing papers explorer
-
Artificial Aphasias in Lesioned Language Models
Lesioning parameters in large language models produces aphasia-like symptoms whose distributions vary by attention versus feed-forward components and by layer depth, but differ qualitatively from human clinical profiles.
-
Hallucination as Commitment Failure: Larger LLMs Misfire Despite Knowing the Answer
Larger LLMs hallucinate more often despite having the correct concept available because instruction tuning causes probability mass to disperse across alternative surface forms instead of concentrating on one.
-
The Proxy Presumption: From Semantic Embeddings to Valid Social Measures
The paper introduces the Construct Validity Protocol to validate semantic embeddings for social constructs and proposes Counterfactual Neutralization using LLMs to reduce confounding.
- Position: State-of-the-Art Claims Require State-of-the-Art Evidence