Lune

ACL Findings2026

Rational Synthesizers or Heuristic Followers? Analyzing LLMs in RAG-based Question-Answering

Atharv Naphade

Abstract

Retrieval-Augmented Generation (RAG) is the prevailing paradigm for grounding Large Language Models (LLMs), yet the mechanisms governing how models integrate groups of conflicting retrieved evidence remain opaque.Does an LLM answer a certain way because the evidence is factually strong, because of a prior belief, or merely because it is repeated frequently?To answer this, we introduce GroupQA, a curated dataset of 1,635 controversial questions paired with 15,058 diverselysourced evidence documents, annotated for stance and qualitative strength.Through controlled experiments, we characterize grouplevel evidence aggregation dynamics: Paraphrasing an argument can be more persuasive than providing distinct independent support; Models favor evidence presented first rather than last, and Larger models are increasingly resistant to adapt to presented evidence.Additionally, we find that LLM explanations to group-based answers are unfaithful.Together, we show that LLMs behave consistently as vulnerable heuristic followers, with direct implications for improving RAG system design.We release the dataset and code for reproducibility: https://github.com/anaphade/GroupQA

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Knowledge Conflicts for LLMs: A Survey cites Towards Understanding Sycophancy in Language Models
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on10

  1. Efficient Streaming Language Models with Attention SinksGuangxuan Xiao, Yuandong Tian, Beidi Chen, Song Han et al.ICLR 2024 · 1,714 citations
  2. Towards Understanding Sycophancy in Language ModelsMrinank Sharma, Meg Tong, Tomasz Korbak, David Duvenaud et al.ICLR 2024 · 762 citations
  3. Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?Sewon Min, Xinxi Lyu, Ari Holtzman, Mikel Artetxe et al.EMNLP 2022 · 634 citations
  4. Making Retrieval-Augmented Language Models Robust to Irrelevant ContextOri Yoran, Tomer Wolfson, Ori Ram, Jonathan BerantICLR 2024 · 361 citations
  5. Knowledge Conflicts for LLMs: A SurveyRongwu Xu, Zehan Qi, Zhijiang Guo, Cunxiang Wang et al.EMNLP 2024 · 38 citations

Related papers

  1. Investigating Context Faithfulness in Large Language Models: The Roles of Memory Strength and Evidence StyleYuepei Li, Kang Zhou, Qiao Qiao, Bach Nguyen et al.ACL Findings 2025 · 5 citations
  2. What Evidence Do Language Models Find Convincing?Alexander Wan, Eric Wallace, Dan KleinACL 2024
  3. Cooperative Retrieval-Augmented Generation for Question Answering: Mutual Information Exchange and Ranking by Contrasting LayersYoumin Ko, Sungjong Seo, Hyunjoon KimNeurIPS 2025 · 2 citations
  4. Intuitive or Dependent? Investigating LLMs' Behavior Style to Conflicting PromptsJiahao Ying, Yixin Cao, Kai Xiong, Long Cui et al.ACL 2024
  5. Faithfulness-Aware Uncertainty Quantification for Fact-Checking the Output of Retrieval-Augmented GenerationEkaterina Fadeeva, Aleksandr Rubashevskii, Dzianis Piatrashyn, Roman Vashurin et al.ACL Findings 2026 · 2 citations
Dusk over the sea between two cliffs drawn in fine vertical lines