Newest papers

Circuits of LLM Classifiers

Hadi Khalaf, Sajani Vithana, Kunal Talreja, Haewon Jeong, Viveck R. Cadambe, Flavio Calmon · 2026 · arXiv

DOI: 10.48550/arXiv.2610.30076 · open-access PDF

Spanish

Los modelos de lenguaje facilitan la construcción de clasificadores al simplemente instar a un modelo con la muestra a clasificar. Sin embargo, sus predicciones no son ni deterministas ni uniformemente fiables en todas las muestras. ¿Cómo podemos combinar predicciones de diferentes modelos para mejorar la fiabilidad? Consideramos los clasificadores LLM como instrumentos de medición ruidosos de una etiqueta objetivo subyacente. Nuestro objetivo es maximizar la exactitud media de clasificación a un coste promedio de consulta dado. A este fin, introducimos circuitos que organizan los clasificadores LLM en una secuencia adaptativa de consultas y decisiones. Después caracterizamos el óptimo

Machine translation (Meta M2M100); the English original follows.

Abstract

Language models make it easy to build classifiers by simply prompting a model with the sample to classify. However, their predictions are neither deterministic nor uniformly reliable across samples. How can we combine predictions from different models to improve reliability? We view LLM classifiers as noisy measurement instruments of an underlying target label. Our goal is to maximize average classification accuracy at a given average query cost. To this end, we introduce circuits that organize LLM classifiers into an adaptive sequence of queries and decisions. We then characterize the optimal

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.30076, 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.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange