Shihui Li, Mette Andersen, Martha Lewis · 2026 · arXiv (Cornell University)
DOI: 10.48550/arxiv.2610.31520 · cited by 0 works (OpenAlex) · open-access PDF
La metáfora es un aspecto crucial de la cognición humana. Como tal, nos gustaría entender cómo se representa en los grandes modelos de lenguaje (LLM) y cómo se puede controlar. Utilizamos técnicas de interpretación mecánica para identificar subespacios causales en una variedad de LLM que cubren diferentes arquitecturas y tamaños. Mostramos que somos capaces de orientar la salida en tareas de clasificación en diferentes partes del habla. También somos capaces de orientar la salida en tareas de generación, indicando que estas técnicas se pueden utilizar para controlar la metafórica de la salida de LLM.
Machine translation (Meta M2M100); the English original follows.
Metaphor is a crucial aspect of human cognition. As such, we would like to understand how it is represented in large language models (LLMs) and how it might be controlled. We use mechanistic interpretability techniques to identify causal subspaces in a range of LLMs covering different architectures and sizes. We show that we are able to steer output on classification tasks across different parts of speech. We are also able to steer output on generation tasks, indicating that these techniques can be used to control the metaphoricity of LLM output.
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
Topics: Interpretability, Metaphor, Computer science, Control (management), Artificial intelligence, Range (aeronautics)
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.31520, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange