arXiv Computation and Language October 21, 2026 By Pratik Worah, Randall Bateman

Can Language Models Get "Aphasia" From Neuronal Ablation?

Read the original on arXiv Computation and Language →

The study investigates how progressive neuronal loss affects language models by systematically ablating computational units in LSTM and Transformer architectures. LSTMs show a gradual decline in syntactic and semantic abilities, while Transformers exhibit a threshold-like collapse under MLP ablation and a smoother sigmoid-like decline under attention-head ablation. These distinct failure modes mirror different aphasic syndromes, suggesting that architecture-dependent degradation patterns can provide computational insights into neurodegenerative diseases.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

arXiv Machine Learning

Aug 14

Perturbation-based Regional Interpretability through Subtraction Mapping (PRISM): naming-error dissociations in language models and post-stroke aphasia

arXiv:2608. 12717v1 Announce Type: new Abstract: Mechanistic interpretability of large language models lacks spatially resolved, falsifiable tools for testing whether internal components are specialized for distinct cognitive operations.

By Xiang Guan, Roger D. Newman-Norlund, Yong Yang, Saeed Ahmadi, Regan Willis, Nadra Salman, Kalil Warren, Srihari Nelakuditi, Chris Rorden, Leonardo Bonilha, Julius Fridriksson

llms safety

arXiv Machine Learning

Aug 10

Recovering Lesion Parameters from Aphasic Picture Naming Error Profiles in Large Language Models

arXiv:2608. 06429v1 Announce Type: cross Abstract: Interpretability methods for large language models (LLMs) describe internal state but do not directly test whether that state is causally sufficient to produce the observed behavior.

By Yong Yang, Roger Newman-Norlund, Xiang Guan, Saeed Ahmadi, Regan Willis, Nadra Salman, Kalil Warren, Sophie Arheix-Parras, Srihari Nelakuditi, Leonardo Bonilha, Christopher Rorden, Rutvik H. Desai, Julius Fridriksson

llms safety

arXiv AI

Jun 9

Component Ablation for Efficient Hybrid Language Model Architectures: Performance, Resilience, and Compression Implications

arXiv:2603. 22473v2 Announce Type: replace-cross Abstract: Hybrid language models combine softmax attention with linear-time sequence mechanisms such as state-space or linear-attention layers, but the functional contribution of each component type remains insufficiently characterized.

By Hector Borobia, Elies Segu\'i-Mas, Guillermina Tormo-Carb\'o

llms benchmarks

arXiv Computation and Language

Oct 22

Language Impairment as a Lens onto Large Language Models

The paper introduces a mechanistic interpretability framework inspired by aphasia to probe large language models (LLMs). Using classifier‑guided lesioning and continuous activation steering, the authors induce aphasia‑like breakdowns and find that language functions can be isolated to discrete structural circuits and continuous vectors, yet these functions are densely entangled with significant overlap across different aphasic profiles. While steering generalizes across downstream tasks, structural lesioning shows asymmetric constraints, highlighting the densely entangled architecture of LLMs that can be mapped through clinically‑informed interventions.

By Muxuan He, Wang Bill Zhu, Yue Chen, Elsi Kaiser, Weizhi Du, Yao Du, Samuel Nastase, Yanjun Gao

llms safety

arXiv AI

Aug 11

LLMs Remember First, Forget Last: Dual-Process Interference in Large Language Models

arXiv:2603. 00270v3 Announce Type: replace-cross Abstract: Large language models can process millions of tokens, yet how they handle conflicting information within context remains poorly understood.

By Sourav Chattaraj, Kanak Raj

llms safety

arXiv AI

Jul 14

Lesioned Multimodal Language Models Reproduce Aphasic Picture-Naming Patterns

arXiv:2607. 11621v1 Announce Type: new Abstract: Aphasia following stroke commonly produces systematic naming errors with characteristic profiles, but whether general-purpose language models not designed for clinical simulation can reproduce these patterns remains untested.

By Yong Yang, Xiang Guan, Sophie Arheix-Parras, Saeed Ahmadi, Roger Newman-Norlund, Leonardo Bonilha, Christopher Rorden, Julius Fridriksson, Rutvik H. Desai, Srihari Nelakuditi

llms multimodal

One email a morning, machine-written

Related stories

Perturbation-based Regional Interpretability through Subtraction Mapping (PRISM): naming-error dissociations in language models and post-stroke aphasia

Recovering Lesion Parameters from Aphasic Picture Naming Error Profiles in Large Language Models

Component Ablation for Efficient Hybrid Language Model Architectures: Performance, Resilience, and Compression Implications

Language Impairment as a Lens onto Large Language Models

LLMs Remember First, Forget Last: Dual-Process Interference in Large Language Models

Lesioned Multimodal Language Models Reproduce Aphasic Picture-Naming Patterns