# Reasoning Isn't Enough: Examining Truth-Bias and Sycophancy in LLMs

arXiv: https://arxiv.org/abs/2506.21561 · HTML: https://mlsanity.com/paper/2506.21561/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning.
- Builds on 1 of its 25 references that are in the map.
- First posted on arXiv 12 Jun 2025.

## Placed in

- LLM reasoning / [Prompting and chain-of-thought reasoning](https://mlsanity.com/lr/prompting-and-chain-of-thought-reasoning/)

## Builds on (most cited first, 1 of 1)

- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)

## Built on by (most cited first, 11 of 11)

- [When Identity Skews Debate: Anonymization for Bias-Reduced Multi-Agent Reasoning](https://mlsanity.com/paper/2510.07517/) (2025)
- [What Counts as AI Sycophancy? A Taxonomy and Expert Survey of a Fragmented Construct](https://mlsanity.com/paper/2605.21778/) (2026)
- [Reward Modeling for Reinforcement Learning-Based LLM Reasoning: Design, Challenges, and Evaluation](https://mlsanity.com/paper/2602.09305/) (2026)
- [Pressure, What Pressure? Sycophancy Disentanglement in Language Models via Reward Decomposition](https://mlsanity.com/paper/2604.05279/) (2026)
- [User-Assistant Bias in LLMs](https://mlsanity.com/paper/2508.15815/) (2025)
- [Getting out of the Big-Muddy: Escalation of Commitment in LLMs](https://mlsanity.com/paper/2508.01545/) (2025)
- [Distortion Instead of Hallucination: The Effect of Reasoning Under Strict Constraints](https://mlsanity.com/paper/2601.01490/) (2026)
- [Evaluating Reasoning Models for Queries with Presuppositions](https://mlsanity.com/paper/2605.03050/) (2026)
- [Needle in the Web: A Benchmark for Retrieving Targeted Web Pages in the Wild](https://mlsanity.com/paper/2512.16553/) (2025)
- [Status Hierarchies in Language Models](https://mlsanity.com/paper/2601.17577/) (2026)
- [When Helpfulness Becomes Sycophancy: Sycophancy is a Boundary Failure Between Social Alignment and Epistemic Integrity in Large Language Models](https://mlsanity.com/paper/2605.05403/) (2026)

## Abstract (arXiv)

Despite their widespread use in fact-checking, moderation, and high-stakes decision-making, large language models (LLMs) remain poorly understood as judges of truth. This study presents the largest evaluation to date of LLMs' veracity detection capabilities and the first analysis of these capabilities in reasoning models. We had eight LLMs make 4,800 veracity judgments across several prompts, comparing reasoning and non-reasoning models. We find that rates of truth-bias, or the likelihood to believe a statement is true, regardless of whether it is actually true, are lower in reasoning models than in non-reasoning models, but still higher than human benchmarks. Most concerning, we identify sycophantic tendencies in several advanced models (o4-mini and GPT-4.1 from OpenAI, R1 from DeepSeek), which displayed an asymmetry in detection accuracy, performing well in truth accuracy but poorly in deception accuracy. This suggests that capability advances alone do not resolve fundamental veracity detection challenges in LLMs.