Newest papers

Effects of Model Scale and Quantization on Social Bias and Abstention in Question Answering

Kuniko Paxton, Ifeoluwa Wuraola, Dhavalkumar Thakker · 2026 · arXiv

DOI: 10.48550/arXiv.2610.29657 · open-access PDF

German

Die Implementierung von kleinen und quantifizierten Sprachmodellen in Ressourcenbeschränkten Einstellungen wirft eine wichtige Fairness-Frage auf: Wenn ein komprimiertes Modell weniger voreingenommene Antworten erzeugt, spiegelt dies ein gerechteres Verhalten oder eine größere Abstinenz wider?Wir untersuchen dieses Problem in Question Answering (QA) unter Verwendung des BBQ-Benchmarks in mehreren Modellgrößen und Quantifizierungs-Einstellungen von Qwen3, Gemma-3 und SmolLM2. Wir untersuchen auch, ob die beobachteten Trends in alternativen Anrufformulationen stabil bleiben und wenn die Option "Unknown" in einer Zwangswahl-Einstellung entfernt wird.

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

Abstract

The deployment of small and quantized language models in resource-constrained settings raises an important fairness question: when a compressed model produces fewer biased responses, does this reflect fairer behavior or greater abstention? We study this problem in Question Answering (QA) using the BBQ benchmark across multiple model sizes and quantization settings from Qwen3, Gemma-3, and SmolLM2. We also examine whether the observed trends remain stable across alternative prompt formulations and when the ``Unknown'' option is removed in a forced-choice setting. Our results show that larger mo

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=2610.29657, 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