paper-with-me

Papers

Safer-Instruct: Aligning Language Models with Automated Preference Data

2023-11-15 · Taiwei Shi, Kai Chen, Jieyu Zhao

Reinforcement learning from human feedback (RLHF) is a vital strategy for enhancing model capability in language models. However, annotating preference data for RLHF is a resource-intensive and creativity-demanding process, while existing automatic generation methods face limitations in data diversity and quality. In response, we present Safer-Instruct, a novel pipeline for automatically constructing large-scale preference data. Our approach leverages reversed instruction tuning, instruction induction, and expert model evaluation to efficiently generate high-quality preference data without human annotators. To verify the effectiveness of Safer-Instruct, we apply the pipeline to construct a safety preference dataset as a case study. Finetuning an Alpaca model on this synthetic dataset not only demonstrates improved harmlessness but also outperforms models fine-tuned on human-annotated safety preference data, all the while maintaining a competitive edge in downstream tasks. Importantly, our Safer-Instruct framework is versatile and can be applied to generate preference data across various domains, extending its utility beyond safety preferences. It addresses the challenges in preference data acquisition and advances the development of more capable and responsible AI systems. For dataset and code implementation, see https://github.com/uscnlp-lime/safer-instruct

📄 PDF Abstract BibTeX arXiv:2311.08685

Code (1)

uscnlp-lime/safer-instruct 공식 구현

Tasks

Diversity

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Adam 설명 없음
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Position-Wise Feed-Forward Layer 설명 없음
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…

Similar Papers 제목 키워드 기반

CodeUltraFeedback: An LLM-as-a-Judge Dataset for Aligning Large Language Models to Coding Preferences

2024-03-14 · Martin Weyssow, Aton Kamanda, Xin Zhou, Houari Sahraoui

Evaluating the alignment of large language models (LLMs) with user-defined coding preferences is a challenging endeavour that requires a deep assessment of LLMs' outputs. Existing methods and benchmarks rely primarily on…

HumanEval

Comparing Bad Apples to Good Oranges: Aligning Large Language Models via Joint Preference Optimization

2024-03-31 · Hritik Bansal, Ashima Suvarna, Gantavya Bhatt, Nanyun Peng 외

A common technique for aligning large language models (LLMs) relies on acquiring human preferences by comparing multiple generations conditioned on a fixed context. This method, however, relies solely on pairwise compari…

SAFER: Probing Safety in Reward Models with Sparse Autoencoder

2025-07-01 · Wei Shi, Ziyuan Xie, Sihang Li, Xiang Wang arxiv

Reinforcement learning from human feedback (RLHF) is a key paradigm for aligning large language models (LLMs) with human values, yet the reward models at its core remain largely opaque. In this work, we present Sparse Au…

Reinforcement Learning

Implicit Cross-Lingual Rewarding for Efficient Multilingual Preference Alignment

2025-03-06 · Wen Yang, Junhong Wu, Chen Wang, Chengqing Zong 외

Direct Preference Optimization (DPO) has become a prominent method for aligning Large Language Models (LLMs) with human preferences. While DPO has enabled significant progress in aligning English LLMs, multilingual prefe…

Instruction FollowingTransfer Learning

Enhancing Robotic Manipulation with AI Feedback from Multimodal Large Language Models

2024-02-22 · Jinyi Liu, Yifu Yuan, Jianye Hao, Fei Ni 외

Recently, there has been considerable attention towards leveraging large language models (LLMs) to enhance decision-making processes. However, aligning the natural language text instructions generated by LLMs with the ve…

Decision MakingRobot Manipulation