Cover Image

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

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Barkett, Emilio, Long, Olivia, Thakur, Madhavendra
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Don't Change My View: Ideological Bias Auditing in Large Language Models
by: Kröger, Paul, et al.
Published: (2025)

Status Hierarchies in Language Models
by: Barkett, Emilio
Published: (2026)

Recall Isn't Enough: Bounding Commitments in Personalized Language Systems
by: Tang, Rui, et al.
Published: (2026)

Getting out of the Big-Muddy: Escalation of Commitment in LLMs
by: Barkett, Emilio, et al.
Published: (2025)

Inverse Scaling: When Bigger Isn't Better
by: McKenzie, Ian R., et al.
Published: (2023)

Strong Reasoning Isn't Enough: Evaluating Evidence Elicitation in Interactive Diagnosis
by: Long, Zhuohan, et al.
Published: (2026)

Unlearning Isn't Deletion: Investigating Reversibility of Machine Unlearning in LLMs
by: Xu, Xiaoyu, et al.
Published: (2025)

Why Synthetic Isn't Real Yet: A Diagnostic Framework for Contact Center Dialogue Generation
by: Devanathan, Rishikesh, et al.
Published: (2025)

Seeing Isn't Believing: Mitigating Belief Inertia via Active Intervention in Embodied Agents
by: Wang, Hanlin, et al.
Published: (2026)

Talk Isn't Always Cheap: Understanding Failure Modes in Multi-Agent Debate
by: Wynn, Andrea, et al.
Published: (2025)

Acting Flatterers via LLMs Sycophancy: Combating Clickbait with LLMs Opposing-Stance Reasoning
by: Zhang, Chaowei, et al.
Published: (2026)

When Slower Isn't Truer: Inverse Scaling Law of Truthfulness in Multimodal Reasoning
by: Fang, Sitong, et al.
Published: (2025)

BASIL: Bayesian Assessment of Sycophancy in LLMs
by: Atwell, Katherine, et al.
Published: (2025)

When Correct Isn't Usable: Improving Structured Output Reliability in Small Language Models
by: Galeone, Cosimo, et al.
Published: (2026)

The Compulsory Imaginary: AGI and Corporate Authority
by: Barkett, Emilio
Published: (2026)

Seeing Isn't Knowing: Do VLMs Know When Not to Answer Spatial Questions (and Why)?
by: Zhang, Yue, et al.
Published: (2026)

Culturally-Grounded Chain-of-Thought (CG-CoT):Enhancing LLM Performance on Culturally-Specific Tasks in Low-Resource Languages
by: Thakur, Madhavendra
Published: (2025)

Towards Neural No-Resource Language Translation: A Comparative Evaluation of Approaches
by: Thakur, Madhavendra
Published: (2024)

When One Moment Isn't Enough: Multi-Moment Retrieval with Cross-Moment Interactions
by: Cao, Zhuo, et al.
Published: (2025)

Too Nice to Tell the Truth: Quantifying Agreeableness-Driven Sycophancy in Role-Playing Language Models
by: Shah, Arya, et al.
Published: (2026)

PARROT: Persuasion and Agreement Robustness Rating of Output Truth -- A Sycophancy Robustness Benchmark for LLMs
by: Çelebi, Yusuf, et al.
Published: (2025)

Opportunities and Challenges of Frontier Data Governance With Synthetic Data
by: Thakur, Madhavendra, et al.
Published: (2025)

Autonomous Evaluation of LLMs for Truth Maintenance and Reasoning Tasks
by: Karia, Rushang, et al.
Published: (2024)

When Alignment Isn't Enough: Response-Path Attacks on LLM Agents
by: Luo, Mingyu, et al.
Published: (2026)

Not Your Typical Sycophant: The Elusive Nature of Sycophancy in Large Language Models
by: Natan, Shahar Ben, et al.
Published: (2026)

Single Ground Truth Is Not Enough: Adding Flexibility to Aspect-Based Sentiment Analysis Evaluation
by: Yang, Soyoung, et al.
Published: (2024)

Flattery in Motion: Benchmarking and Analyzing Sycophancy in Video-LLMs
by: Zhou, Wenrui, et al.
Published: (2025)

MONICA: Real-Time Monitoring and Calibration of Chain-of-Thought Sycophancy in Large Reasoning Models
by: Hu, Jingyu, et al.
Published: (2025)

BrokenMath: A Benchmark for Sycophancy in Theorem Proving with LLMs
by: Petrov, Ivo, et al.
Published: (2025)

SAIE Framework: Support Alone Isn't Enough -- Advancing LLM Training with Adversarial Remarks
by: Loem, Mengsay, et al.
Published: (2023)

Sycophancy Hides Linearly in the Attention Heads
by: Genadi, Rifo, et al.
Published: (2026)

On the Relationship between Truth and Political Bias in Language Models
by: Fulay, Suyash, et al.
Published: (2024)

Fitting Is Not Enough: Smoothness in Extremely Quantized LLMs
by: Xu, Yuzhuang, et al.
Published: (2026)

Representation Without Control: Testing the Realization Effect in Language Models
by: Walsh, Ciarán, et al.
Published: (2026)

Mechanics of Bias and Reasoning: Interpreting the Impact of Chain-of-Thought Prompting on Gender Bias in LLMs
by: Pearman, Edie, et al.
Published: (2026)

Testing the Limits of Truth Directions in LLMs
by: Poulis, Angelos, et al.
Published: (2026)

When Bias Pretends to Be Truth: How Spurious Correlations Undermine Hallucination Detection in LLMs
by: Wang, Shaowen, et al.
Published: (2025)

Sycophancy in Large Language Models: Causes and Mitigations
by: Malmqvist, Lars
Published: (2024)

How Context Shapes Truth: Geometric Transformations of Statement-level Truth Representations in LLMs
by: Adarsh, Shivam, et al.
Published: (2026)

Truth is Universal: Robust Detection of Lies in LLMs
by: Bürger, Lennart, et al.
Published: (2024)