Image de couverture de livre

Mitigating Sycophancy in Decoder-Only Transformer Architectures: Synthetic Data Intervention

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteur principal:	Wang, Libo
Format:	Preprint
Publié:	2024
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Sycophancy in Large Language Models: Causes and Mitigations
par: Malmqvist, Lars
Publié: (2024)

Sycophancy Mitigation Through Reinforcement Learning with Uncertainty-Aware Adaptive Reasoning Trajectories
par: Beigi, Mohammad, et autres
Publié: (2025)

Benchmarking and Mitigating Sycophancy in Medical Vision Language Models
par: Xu, Juangui, et autres
Publié: (2025)

Diagnosing and Mitigating Sycophancy and Skepticism in LLM Causal Judgment
par: Chang, Edward Y.
Publié: (2026)

EDIT: Enhancing Vision Transformers by Mitigating Attention Sink through an Encoder-Decoder Architecture
par: Feng, Wenfeng, et autres
Publié: (2025)

Towards Humanoid Robot Autonomy: A Dynamic Architecture Integrating Continuous thought Machines (CTM) and Model Context Protocol (MCP)
par: Wang, Libo
Publié: (2025)

Beacon: Single-Turn Diagnosis and Mitigation of Latent Sycophancy in Large Language Models
par: Pandey, Sanskar, et autres
Publié: (2025)

How RLHF Amplifies Sycophancy
par: Shapira, Itai, et autres
Publié: (2026)

Sycophancy in Vision-Language Models: A Systematic Analysis and an Inference-Time Mitigation Framework
par: Zhao, Yunpu, et autres
Publié: (2024)

Transforming Decoder-Only Transformers for Accurate WiFi-Telemetry Based Indoor Localization
par: Bhatia, Nayan Sanjay, et autres
Publié: (2025)

Moral Sycophancy in Vision Language Models
par: Rabby, Shadman, et autres
Publié: (2026)

SycEval: Evaluating LLM Sycophancy
par: Fanous, Aaron, et autres
Publié: (2025)

Sycophancy is an Educational Safety Risk: Why LLM Tutors Need Sycophancy Benchmarks
par: Kasneci, Enkelejda, et autres
Publié: (2026)

Not Your Typical Sycophant: The Elusive Nature of Sycophancy in Large Language Models
par: Natan, Shahar Ben, et autres
Publié: (2026)

Mapping the Edge of Chaos: Fractal-Like Boundaries in The Trainability of Decoder-Only Transformer Models
par: Torkamandi, Bahman
Publié: (2025)

Linear Probe Penalties Reduce LLM Sycophancy
par: Papadatos, Henry, et autres
Publié: (2024)

When Helpfulness Becomes Sycophancy: Sycophancy is a Boundary Failure Between Social Alignment and Epistemic Integrity in Large Language Models
par: Li, Jiechen, et autres
Publié: (2026)

Self-Blinding and Counterfactual Self-Simulation Mitigate Biases and Sycophancy in Large Language Models
par: Christian, Brian, et autres
Publié: (2026)

Indoor Localization using Compact, Telemetry-Agnostic, Transfer-Learning Enabled Decoder-Only Transformer
par: Bhatia, Nayan Sanjay, et autres
Publié: (2025)

BASIL: Bayesian Assessment of Sycophancy in LLMs
par: Atwell, Katherine, et autres
Publié: (2025)

Sycophancy Hides Linearly in the Attention Heads
par: Genadi, Rifo, et autres
Publié: (2026)

The Hidden Cost of Contextual Sycophancy: an AI Literacy Intervention in Human-AI Collaboration
par: Koyuturk, Cansu, et autres
Publié: (2026)

Task Decoding based on Eye Movements using Synthetic Data Augmentation
par: Sadhu, Shanmuka, et autres
Publié: (2025)

On the Effectiveness of Acoustic BPE in Decoder-Only TTS
par: Li, Bohan, et autres
Publié: (2024)

SynEHRgy: Synthesizing Mixed-Type Structured Electronic Health Records using Decoder-Only Transformers
par: Karami, Hojjat, et autres
Publié: (2024)

Political Bias Audits of LLMs Capture Sycophancy to the Inferred Auditor
par: Törnberg, Petter, et autres
Publié: (2026)

Consistency Training Helps Stop Sycophancy and Jailbreaks
par: Irpan, Alex, et autres
Publié: (2025)

Trained Persistent Memory for Frozen Decoder-Only LLMs
par: Jeong, Hong
Publié: (2026)

Towards Better Code Understanding in Decoder-Only Models with Contrastive Learning
par: Lin, Jiayi, et autres
Publié: (2024)

Synthetic4Health: Generating Annotated Synthetic Clinical Letters
par: Ren, Libo, et autres
Publié: (2024)

PENDULUM: A Benchmark for Assessing Sycophancy in Multimodal Large Language Models
par: Rahman, A. B. M. Ashikur, et autres
Publié: (2025)

Equilibrium Dynamics and Mitigation of Gender Bias in Synthetically Generated Data
par: Kattamuri, Ashish, et autres
Publié: (2025)

Multi-Scenario Reasoning: Unlocking Cognitive Autonomy in Humanoid Robots for Multimodal Understanding
par: Wang, Libo
Publié: (2024)

Outcome-Based RL Provably Leads Transformers to Reason, but Only With the Right Data
par: Ran-Milo, Yuval, et autres
Publié: (2026)

Pressure, What Pressure? Sycophancy Disentanglement in Language Models via Reward Decomposition
par: Mohsin, Muhammad Ahmed, et autres
Publié: (2026)

Set Transformer Architectures and Synthetic Data Generation for Flow-Guided Nanoscale Localization
par: Hube, Mika Leo, et autres
Publié: (2025)

Dynamic Chain-of-Thought: Towards Adaptive Deep Reasoning
par: Wang, Libo
Publié: (2025)

Scaling Laws of Decoder-Only Models on the Multilingual Machine Translation Task
par: Caillaut, Gaëtan, et autres
Publié: (2024)

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models
par: Denison, Carson, et autres
Publié: (2024)

The Price of Agreement: Measuring LLM Sycophancy in Agentic Financial Applications
par: Zhao, Zhenyu, et autres
Publié: (2026)