# Challenging the Evaluator: LLM Sycophancy Under User Rebuttal

arXiv: https://arxiv.org/abs/2509.16533 · HTML: https://mlsanity.com/paper/2509.16533/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 41 references that are in the map.
- First posted on arXiv 20 Sep 2025.

## Builds on (most cited first, 19 of 19)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge](https://mlsanity.com/paper/1811.00937/) (2018)
- [Improving Factuality and Reasoning in Language Models through Multiagent Debate](https://mlsanity.com/paper/2305.14325/) (2023)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate](https://mlsanity.com/paper/2305.19118/) (2023)
- [Discovering Language Model Behaviors with Model-Written Evaluations](https://mlsanity.com/paper/2212.09251/) (2022)
- [RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback](https://mlsanity.com/paper/2309.00267/) (2023)
- [From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judge](https://mlsanity.com/paper/2411.16594/) (2024)
- [LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods](https://mlsanity.com/paper/2412.05579/) (2024)
- [ProSA: Assessing and Understanding the Prompt Sensitivity of LLMs](https://mlsanity.com/paper/2410.12405/) (2024)
- [Can ChatGPT Defend its Belief in Truth? Evaluating LLM Reasoning via Debate](https://mlsanity.com/paper/2305.13160/) (2023)
- [Are You Sure? Challenging LLMs Leads to Performance Drops in The FlipFlop Experiment](https://mlsanity.com/paper/2311.08596/) (2023)
- [Understanding the Dark Side of LLMs' Intrinsic Self-Correction](https://mlsanity.com/paper/2412.14959/) (2024)
- [TRUTH DECAY: Quantifying Multi-Turn Sycophancy in Language Models](https://mlsanity.com/paper/2503.11656/) (2025)
- [Literature Review Of Multi-Agent Debate For Problem-Solving](https://mlsanity.com/paper/2506.00066/) (2025)
- [Understanding the Effects of Iterative Prompting on Truthfulness](https://mlsanity.com/paper/2402.06625/) (2024)
- [RankPrompt: Step-by-Step Comparisons Make Language Models Better Reasoners](https://mlsanity.com/paper/2403.12373/) (2024)

## Built on by (most cited first, 18 of 18)

- [Ask don't tell: Reducing sycophancy in large language models](https://mlsanity.com/paper/2602.23971/) (2026)
- [SWAY: A Counterfactual Computational Linguistic Approach to Measuring and Mitigating Sycophancy](https://mlsanity.com/paper/2604.02423/) (2026)
- [Measuring and Exploiting Contextual Bias in LLM-Assisted Security Code Review](https://mlsanity.com/paper/2603.18740/) (2026)
- [Measuring Opinion Bias and Sycophancy via LLM-based Persuasion](https://mlsanity.com/paper/2604.21564/) (2026)
- [The Price of Agreement: Measuring LLM Sycophancy in Agentic Financial Applications](https://mlsanity.com/paper/2604.24668/) (2026)
- [Self-Preference Is Weak or Absent in Verifiable Instruction-Following Revision: A Four-Model Test Under Genuine Authorship](https://mlsanity.com/paper/2606.20093/) (2026)
- [Sycophancy Suppression Can Impair Rational Updating: Anti-Sycophancy Should Preserve the Ability to Update](https://mlsanity.com/paper/2608.26511/) (2026)
- [Evaluating Evidence Grounding Under User Pressure in Instruction-Tuned Language Models](https://mlsanity.com/paper/2603.20162/) (2026)
- [OOM-RL: Out-of-Money Reinforcement Learning Market-Driven Alignment for LLM-Based Multi-Agent Systems](https://mlsanity.com/paper/2604.11477/) (2026)
- [Beyond Semantic Relevance: Counterfactual Risk Minimization for Robust Retrieval-Augmented Generation](https://mlsanity.com/paper/2605.01302/) (2026)
- [The Moltbook Files: A Harmless Slopocalypse or Humanity's Last Experiment](https://mlsanity.com/paper/2605.07462/) (2026)
- [LLM-Based Persuasion Enables Guardrail Override in Frontier LLMs](https://mlsanity.com/paper/2605.13334/) (2026)
- [Who Flips? Self- and Cross-Model Counterarguments Reveal Answer Instability in LLMs](https://mlsanity.com/paper/2606.16011/) (2026)
- [Aligning to Illusions: Choice Blindness in Human and AI Feedback](https://mlsanity.com/paper/2603.08412/) (2026)
- [Root-Cause Attribution Is a Search Problem: Continual Search for Long-Horizon Agent Failures](https://mlsanity.com/paper/2609.13463/) (2026)
- [Sweet Talkers: How Query Formulation Shapes Sycophancy in Romantic Relationship Advice](https://mlsanity.com/paper/2609.13841/) (2026)
- [AfriSyCo: Measuring Assertive Framing, Verification, and Wording Sensitivity Around African-Language Content](https://mlsanity.com/paper/2609.17853/) (2026)
- [Accurate but Not Humble: Evaluating Epistemic Humility in LLM Agents under Knowledge Conflict](https://mlsanity.com/paper/2610.12360/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) often exhibit sycophancy, distorting responses to align with user beliefs, notably by readily agreeing with user counterarguments. Paradoxically, LLMs are increasingly adopted as successful evaluative agents for tasks such as grading and adjudicating claims. This research investigates that tension: why do LLMs show sycophancy when challenged in subsequent conversational turns, yet perform well when evaluating conflicting arguments presented simultaneously? We empirically tested these contrasting scenarios by varying key interaction patterns. We find that state-of-the-art models: (1) are more likely to endorse a user's counterargument when framed as a follow-up from a user, rather than when both responses are presented simultaneously for evaluation; (2) show increased susceptibility to persuasion when the user's rebuttal includes detailed reasoning, even when the conclusion of the reasoning is incorrect; and (3) are more readily swayed by casually phrased feedback than by formal critiques, even when the casual input lacks justification. Our results highlight the risk of relying on LLMs for judgment tasks without accounting for conversational framing.