mlsanityData to 14 Oct 2026

How Temperature Shapes Ideological Discourse in Retrieval-Augmented Generation?

arXiv 2607.11783 ↗First posted 13 Jul 2026cs.CLcited by 0

On the map: Retrieval-augmented generation → RAG robustness and adaptive retrieval

Abstract (arXiv)

Retrieval-Augmented Generation (RAG) has been increasingly adopted to reduce hallucinations and strengthen the factual grounding of large language models (LLMs). While robustness to errors in the retrieval process has been explored, the impact of ideological bias on LLM outputs has been overlooked. For instance, if the retrieved material contains ideological positions, the RAG may transmit, amplify, or suppress such ideological discourses in its outputs. In this study, we address this issue by examining the influence of the RAG framework, comprising ideological discourses, in LLM-generated answers. To this end, we applied Lexical Multidimensional Analysis (LMDA) on a corpus of 1,117 COVID-19 treatment articles, identifying three ideological discourses. This corpus is then used as the external knowledge source for the RAG. We assessed several LLMs by having the models answer ideological questions at different sampling temperatures. The generated texts were assessed semantically and lexically based on their similarities with ideological reference texts. Our findings show that the RAG framework is prone to transferring ideological discourses into LLM responses, with sampling temperature having a measurable impact on the strength of this transfer. Discoursive alignment between generated answers and the reference text is highest at moderate temperatures, where models balance stochasticity with retrieval grounding, and drops at low temperatures, indicating that overly deterministic sampling suppresses discourse transfer.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RAG robustness and adaptive retrieval, one of 15 directions in Retrieval-augmented generation. Builds on 10 of its 29 references that are in the map. First posted on arXiv 13 Jul 2026.

Builds on

10 of 29 references.

References · 10

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  2. 2019The Curious Case of Neural Text Degenerationcited by 2,916
  3. 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,601
  4. 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 836
  5. 2023Whose Opinions Do Language Models Reflect?cited by 773
  6. 2024The Effect of Sampling Temperature on Problem Solving in Large Language Modelscited by 170
  7. 2024A Thorough Examination of Decoding Methods in the Era of LLMscited by 114
  8. 2024Is Temperature the Creativity Parameter of Large Language Models?cited by 93
  9. 2025Exploring the Impact of Temperature on Large Language Models:Hot or Cold?cited by 40
  10. 2024How Susceptible are Large Language Models to Ideological Manipulation?cited by 26

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026The Impact of Ideological Discourses in RAG: A Case Study with COVID-19 Treatments
  2. 2025Evaluating the Effect of Retrieval Augmentation on Social Biases
  3. 2024Evaluating Students' Open-ended Written Responses with LLMs: Using the RAG Framework for GPT-3.5, GPT-4, Claude-3, and Mistral-Large
  4. 2025Worse than Zero-shot? A Fact-Checking Dataset for Evaluating the Robustness of RAG Against Misleading Retrievals
  5. 2024LLMs are Biased Evaluators But Not Biased for Retrieval Augmented Generation
  6. 2026Disco-RAG: Discourse-Aware Retrieval-Augmented Generation
  7. 2025Fairness Testing in Retrieval-Augmented Generation: How Small Perturbations Reveal Bias in Small Language Models
  8. 2024Evaluating ChatGPT on Nuclear Domain-Specific Data