paper-with-me

홈 › Papers

Training Language Models with Language Feedback

2022-04-29 · Jérémy Scheurer, Jon Ander Campos, Jun Shern Chan, Angelica Chen, Kyunghyun Cho, Ethan Perez

Pretrained language models often do not perform tasks in ways that are in line with our preferences, e.g., generating offensive text or factually incorrect summaries. Recent work approaches the above issue by learning from a simple form of human evaluation: comparisons between pairs of model-generated task outputs. Comparison feedback conveys limited information about human preferences per human evaluation. Here, we propose to learn from natural language feedback, which conveys more information per human evaluation. We learn from language feedback on model outputs using a three-step learning algorithm. First, we condition the language model on the initial output and feedback to generate many refinements. Second, we choose the refinement with the highest similarity to the feedback. Third, we finetune a language model to maximize the likelihood of the chosen refinement given the input. In synthetic experiments, we first evaluate whether language models accurately incorporate feedback to produce refinements, finding that only large language models (175B parameters) do so. Using only 100 samples of human-written feedback, our learning algorithm finetunes a GPT-3 model to roughly human-level summarization ability.

📄 PDF Abstract BibTeX arXiv:2204.14146

Code (0)

등록된 구현이 없습니다.

Tasks

Language ModelingLanguage Modelling

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Multi-Head Attention 설명 없음
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Cosine Annealing Cosine Annealing is a type of learning rate schedule that has the effect of starting with a large learning rate that is relatively rapidly decreased to a minimum value before…
Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…

Similar Papers 제목 키워드 기반

Bridging the Gap: A Survey on Integrating (Human) Feedback for Natural Language Generation

2023-05-01 · Patrick Fernandes, Aman Madaan, Emmy Liu, António Farinhas 외

Many recent advances in natural language generation have been fueled by training large language models on internet-scale data. However, this paradigm can lead to models that generate toxic, inaccurate, and unhelpful cont…

Text Generation

Improving Code Generation by Training with Natural Language Feedback

2023-03-28 · Angelica Chen, Jérémy Scheurer, Tomasz Korbak, Jon Ander Campos 외

The potential for pre-trained large language models (LLMs) to use natural language feedback at inference time has been an exciting recent development. We build upon this observation by formalizing an algorithm for learni…

Code GenerationImitation LearningmbppProgram Synthesis

Is Feedback All You Need? Leveraging Natural Language Feedback in Goal-Conditioned Reinforcement Learning

2023-12-07 · Sabrina McCallum, Max Taylor-Davies, Stefano V. Albrecht, Alessandro Suglia

Despite numerous successes, the field of reinforcement learning (RL) remains far from matching the impressive generalisation power of human behaviour learning. One possible way to help bridge this gap be to provide RL ag…

AllReinforcement Learning (RL)

Training Language Models with Language Feedback at Scale

2023-03-28 · Jérémy Scheurer, Jon Ander Campos, Tomasz Korbak, Jun Shern Chan 외

Pretrained language models often generate outputs that are not in line with human preferences, such as harmful text or factually incorrect summaries. Recent work approaches the above issues by learning from a simple form…

Bayesian InferenceImitation LearningLanguage ModelingLanguage Modelling

Constructive Large Language Models Alignment with Diverse Feedback

2023-10-10 · Tianshu Yu, Ting-En Lin, Yuchuan Wu, Min Yang 외

In recent research on large language models (LLMs), there has been a growing emphasis on aligning these models with human values to reduce the impact of harmful content. However, current alignment methods often rely sole…

Learning TheoryModels AlignmentQuestion AnsweringText Summarization