Cover Image

Training language models to be warm and empathetic makes them less reliable and more sycophantic

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Ibrahim, Lujain, Hafner, Franziska Sofia, Rocher, Luc
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Artificial Intelligence Computers and Society
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Sycophantic AI makes human interaction feel more effortful and less satisfying over time
by: Ibrahim, Lujain, et al.
Published: (2026)

Gender Trouble in Language Models: An Empirical Audit Guided by Gender Performativity Theory
by: Hafner, Franziska Sofia, et al.
Published: (2025)

Characterizing and modeling harms from interactions with design patterns in AI interfaces
by: Ibrahim, Lujain, et al.
Published: (2024)

Developmental trajectories of decision making and affective dynamics in large language models
by: Wang, Zhihao, et al.
Published: (2025)

ELEPHANT: Measuring and understanding social sycophancy in LLMs
by: Cheng, Myra, et al.
Published: (2025)

Verbalizing LLMs' assumptions to explain and control sycophancy
by: Cheng, Myra, et al.
Published: (2026)

Post-training makes large language models less human-like
by: Binz, Marcel, et al.
Published: (2026)

Uncovering inequalities in new knowledge learning by large language models across different languages
by: Wang, Chenglong, et al.
Published: (2025)

Retrieval-augmented reasoning with lean language models
by: Chan, Ryan Sze-Yin, et al.
Published: (2025)

Do Chinese models speak Chinese languages?
by: Wen-Yi, Andrea W, et al.
Published: (2025)

Failure of contextual invariance in large language models
by: Kumar, Sagar, et al.
Published: (2026)

Large language models in medicine: the potentials and pitfalls
by: Omiye, Jesutofunmi A., et al.
Published: (2023)

What can large language models do for sustainable food?
by: Thomas, Anna T., et al.
Published: (2025)

A closer look at how large language models trust humans: patterns and biases
by: Lerman, Valeria, et al.
Published: (2025)

A survey on fairness of large language models in e-commerce: progress, application, and challenge
by: Ren, Qingyang, et al.
Published: (2024)

Large language models can consistently generate high-quality content for election disinformation operations
by: Williams, Angus R., et al.
Published: (2024)

AI-AI Bias: large language models favor communications generated by large language models
by: Laurito, Walter, et al.
Published: (2024)

Redefining technology for indigenous languages
by: Fernandez-Sabido, Silvia, et al.
Published: (2025)

Meaning Is Not A Metric: Using LLMs to make cultural context legible at scale
by: Kommers, Cody, et al.
Published: (2025)

Can LLMs make trade-offs involving stipulated pain and pleasure states?
by: Keeling, Geoff, et al.
Published: (2024)

Can adversarial attacks by large language models be attributed?
by: Cebrian, Manuel, et al.
Published: (2024)

HumT DumT: Measuring and controlling human-like language in LLMs
by: Cheng, Myra, et al.
Published: (2025)

Developing Story: Case Studies of Generative AI's Use in Journalism
by: Brigham, Natalie Grace, et al.
Published: (2024)

Implicit assessment of language learning during practice as accurate as explicit testing
by: Hou, Jue, et al.
Published: (2024)

Profiling learners' affective engagement: Emotion AI, intercultural pragmatics, and language learning
by: Godwin-Jones, Robert
Published: (2026)

Assessing the nature of large language models: A caution against anthropocentrism
by: Speed, Ann
Published: (2023)

A validity-guided workflow for robust large language model research in psychology
by: Lin, Zhicheng
Published: (2025)

Evidence of a log scaling law for political persuasion with large language models
by: Hackenburg, Kobi, et al.
Published: (2024)

Towards interactive evaluations for interaction harms in human-AI systems
by: Ibrahim, Lujain, et al.
Published: (2024)

What are human values, and how do we align AI to them?
by: Klingefjord, Oliver, et al.
Published: (2024)

Human Decision-making is Susceptible to AI-driven Manipulation
by: Sabour, Sahand, et al.
Published: (2025)

WiseMind: a knowledge-guided multi-agent framework for accurate and empathetic psychiatric diagnosis
by: Wu, Yuqi, et al.
Published: (2025)

What Is The Political Content in LLMs' Pre- and Post-Training Data?
by: Ceron, Tanise, et al.
Published: (2025)

The Last Fingerprint: How Markdown Training Shapes LLM Prose
by: Freeburg, E. M.
Published: (2026)

PersLLM: A Personified Training Approach for Large Language Models
by: Zeng, Zheni, et al.
Published: (2024)

From Hard Refusals to Safe-Completions: Toward Output-Centric Safety Training
by: Yuan, Yuan, et al.
Published: (2025)

Training-Free Cultural Alignment of Large Language Models via Persona Disagreement
by: Kiet, Huynh Trung, et al.
Published: (2026)

The opportunities and risks of large language models in mental health
by: Lawrence, Hannah R., et al.
Published: (2024)

Readers Prefer Outputs of AI Trained on Copyrighted Books over Expert Human Writers
by: Chakrabarty, Tuhin, et al.
Published: (2025)

Can a large language model be a gaslighter?
by: Li, Wei, et al.
Published: (2024)