pith. sign in

arxiv: 2408.07852 · v1 · pith:M7VPP2ANnew · submitted 2024-08-14 · 💻 cs.CL · cs.AI· cs.LG

Training Language Models on the Knowledge Graph: Insights on Hallucinations and Their Detectability

classification 💻 cs.CL cs.AIcs.LG
keywords hallucinationstrainingscaledatadatasetdetectabilityfindfixed
0
0 comments X
read the original abstract

While many capabilities of language models (LMs) improve with increased training budget, the influence of scale on hallucinations is not yet fully understood. Hallucinations come in many forms, and there is no universally accepted definition. We thus focus on studying only those hallucinations where a correct answer appears verbatim in the training set. To fully control the training data content, we construct a knowledge graph (KG)-based dataset, and use it to train a set of increasingly large LMs. We find that for a fixed dataset, larger and longer-trained LMs hallucinate less. However, hallucinating on $\leq5$% of the training data requires an order of magnitude larger model, and thus an order of magnitude more compute, than Hoffmann et al. (2022) reported was optimal. Given this costliness, we study how hallucination detectors depend on scale. While we see detector size improves performance on fixed LM's outputs, we find an inverse relationship between the scale of the LM and the detectability of its hallucinations.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Budget-Constrained Online Retrieval-Augmented Generation: The Chunk-as-a-Service Model

    cs.IR 2026-04 unverdicted novelty 5.0

    Chunk-as-a-Service with the UCOSA online algorithm enables budget-constrained selection of prompts for chunk enrichment in RAG, outperforming random selection by 52% on a combined performance metric and delivering hig...