Linear Probe Penalties Reduce LLM Sycophancy

• URL: http://arxiv.org/abs/2412.00967v1
• Date: Sun, 01 Dec 2024 21:11:28 GMT
• Title: Linear Probe Penalties Reduce LLM Sycophancy
• Authors: Henry Papadatos, Rachel Freedman,
• Abstract summary: Large language models (LLMs) are often sycophantic, prioritizing agreement with their users over accurate or objective statements. This problematic behavior becomes more pronounced during reinforcement learning from human feedback (RLHF) We develop a linear probing method to identify and penalize markers of sycophancy within the reward model, producing rewards that discourage sycophantic behavior.
• Score: 3.6490659260835234
• License:
• Abstract: Large language models (LLMs) are often sycophantic, prioritizing agreement with their users over accurate or objective statements. This problematic behavior becomes more pronounced during reinforcement learning from human feedback (RLHF), an LLM fine-tuning stage intended to align model outputs with human values. Instead of increasing accuracy and reliability, the reward model learned from RLHF often rewards sycophancy. We develop a linear probing method to identify and penalize markers of sycophancy within the reward model, producing rewards that discourage sycophantic behavior. Our experiments show that constructing and optimizing against this surrogate reward function reduces sycophantic behavior in multiple open-source LLMs. Our results suggest a generalizable methodology for reducing unwanted LLM behaviors that are not sufficiently disincentivized by RLHF fine-tuning.

Related papers

• Beyond Reward Hacking: Causal Rewards for Large Language Model Alignment [30.605500809158986]
  We propose a novel causal reward modeling approach that integrates causal inference to mitigate spurious correlations. Our approach mitigates various types of spurious correlations effectively, resulting in more reliable and fair alignment of LLMs with human preferences.
  arXiv  Detail & Related papers  (2025-01-16T16:00:37Z)
• Insights from the Inverse: Reconstructing LLM Training Goals Through Inverse RL [7.988692259455583]
  Large language models (LLMs) trained with Reinforcement Learning from Human Feedback have demonstrated remarkable capabilities, but their underlying reward functions and decision-making processes remain opaque. This paper introduces a novel approach to interpreting LLMs by applying inverse reinforcement learning (IRL) to recover their implicit reward functions. We conduct experiments on toxicity-aligned LLMs of varying sizes, extracting reward models that achieve up to 80.40% accuracy in predicting human preferences.
  arXiv  Detail & Related papers  (2024-10-16T12:14:25Z)
• Have the VLMs Lost Confidence? A Study of Sycophancy in VLMs [44.56018149475948]
  sycophancy is a prevalent hallucination that poses significant challenges to visual language models (VLMs) We propose a synthetic dataset for training and employ methods based on prompts, supervised fine-tuning, and DPO to mitigate sycophancy. Our findings indicate that the ability to prevent sycophancy is predominantly observed in higher layers of the model.
  arXiv  Detail & Related papers  (2024-10-15T05:48:14Z)
• Sequence to Sequence Reward Modeling: Improving RLHF by Language Feedback [8.601283886845664]
  Reinforcement learning from human feedback (RLHF) aligns Large language models (LLMs) with human intentions and values. Despite its effectiveness and popularity, RLHF is prone to biased local optimization. We propose a novel textitsequence-to-sequence (seq2seq) reward modeling method.
  arXiv  Detail & Related papers  (2024-08-30T16:14:35Z)
• CoMMIT: Coordinated Instruction Tuning for Multimodal Large Language Models [68.64605538559312]
  In this paper, we analyze the MLLM instruction tuning from both theoretical and empirical perspectives. Inspired by our findings, we propose a measurement to quantitatively evaluate the learning balance. In addition, we introduce an auxiliary loss regularization method to promote updating of the generation distribution of MLLMs.
  arXiv  Detail & Related papers  (2024-07-29T23:18:55Z)
• Self-Exploring Language Models: Active Preference Elicitation for Online Alignment [88.56809269990625]
  We propose a bilevel objective optimistically biased towards potentially high-reward responses to actively explore out-of-distribution regions. Our experimental results demonstrate that when fine-tuned on Zephyr-7B-SFT and Llama-3-8B-Instruct models, Self-Exploring Language Models (SELM) significantly boosts the performance on instruction-following benchmarks.
  arXiv  Detail & Related papers  (2024-05-29T17:59:07Z)
• Countering Reward Over-optimization in LLM with Demonstration-Guided Reinforcement Learning [49.87923965553233]
  Reinforcement Learning can lead to reward over-optimization in large language models. We introduce the Reward from Demonstration (RCfD) to recalibrate the reward objective. We show that RCfD achieves comparable performance to carefully tuned baselines while mitigating ROO.
  arXiv  Detail & Related papers  (2024-04-30T09:57:21Z)
• Reinforcement Learning from Reflective Feedback (RLRF): Aligning and Improving LLMs via Fine-Grained Self-Reflection [24.435121488662897]
  We propose a novel framework: Reinforcement Learning from Reflective Feedback (RLRF) RLRF employs a self-reflection mechanism to systematically explore and refine LLM responses, then fine-tuning the models via a RL algorithm along with promising responses. Our experiments across Just-Eval, Factuality, and Mathematical Reasoning demonstrate the efficacy and transformative potential of RLRF.
  arXiv  Detail & Related papers  (2024-03-21T08:57:27Z)
• DeAL: Decoding-time Alignment for Large Language Models [59.63643988872571]
  Large Language Models (LLMs) are nowadays expected to generate content aligned with human preferences. We propose DeAL, a framework that allows the user to customize reward functions and enables Detime Alignment of LLMs. Our experiments show that we can DeAL with fine-grained trade-offs, improve adherence to alignment objectives, and address residual gaps in LLMs.
  arXiv  Detail & Related papers  (2024-02-05T06:12:29Z)
• RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback [103.08766858584049]
  We present RLHF-V, which enhances MLLM trustworthiness via behavior alignment from fine-grained correctional human feedback. Experiments on five benchmarks in both automatic and human evaluation show that, RLHF-V can enable substantially more trustworthy MLLM behaviors.
  arXiv  Detail & Related papers  (2023-12-01T11:36:08Z)
• Gaining Wisdom from Setbacks: Aligning Large Language Models via Mistake Analysis [127.85293480405082]
  The rapid development of large language models (LLMs) has not only provided numerous opportunities but also presented significant challenges. Existing alignment methods usually direct LLMs toward the favorable outcomes by utilizing human-annotated, flawless instruction-response pairs. This study proposes a novel alignment technique based on mistake analysis, which deliberately exposes LLMs to erroneous content to learn the reasons for mistakes and how to avoid them.
  arXiv  Detail & Related papers  (2023-10-16T14:59:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.