9.7k citations · 11.2k across the 18 of their papers we have counts for
19 papers
Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla
Tom Lieberum, Matthew Rahtz, János Kramár +4
\emph{Circuit analysis} is a promising technique for understanding the internal mechanisms of language models. However, existing analyses are done in small models far from the stat…
Accelerating Large Language Model Decoding with Speculative Sampling
Charlie Chen, Sebastian Borgeaud, Geoffrey Irving +3
We present speculative sampling, an algorithm for accelerating transformer decoding by enabling the generation of multiple tokens from each transformer call. Our algorithm relies o…
Solving math word problems with process- and outcome-based feedback
Jonathan Uesato, Nate Kushman, Ramana Kumar +6
Recent work has shown that asking language models to generate reasoning steps improves performance on many reasoning tasks. When moving beyond prompting, this raises the question o…
Fine-Tuning Language Models via Epistemic Neural Networks
Ian Osband, Seyed Mohammad Asghari, Benjamin Van Roy +3
Language models often pre-train on large unsupervised text corpora, then fine-tune on additional task-specific data. However, typical fine-tuning schemes do not prioritize the exam…
Improving alignment of dialogue agents via targeted human judgements
Amelia Glaese, Nat McAleese, Maja Trębacz +31
We present Sparrow, an information-seeking dialogue agent trained to be more helpful, correct, and harmless compared to prompted language model baselines. We use reinforcement lear…
Characteristics of Harmful Text: Towards Rigorous Benchmarking of Language Models
Maribeth Rauh, John Mellor, Jonathan Uesato +9
Large language models produce human-like text that drive a growing number of applications. However, recent literature and, increasingly, real world observations, have demonstrated…
Teaching language models to support answers with verified quotes
Jacob Menick, Maja Trebacz, Vladimir Mikulik +8
Recent large language models often answer factual questions correctly. But users can't trust any given claim a model makes without fact-checking, because language models can halluc…
Uncertainty Estimation for Language Reward Models
Adam Gleave, Geoffrey Irving
Language models can learn a range of capabilities from unsupervised training on text corpora. However, to solve a particular problem (such as text summarization) it is typically ne…
Red Teaming Language Models with Language Models
Ethan Perez, Saffron Huang, Francis Song +6
Language Models (LMs) often cannot be deployed because of their potential to harm users in hard-to-predict ways. Prior work identifies harmful behaviors before deployment by using…
Scaling Language Models: Methods, Analysis & Insights from Training Gopher
Jack W. Rae, Sebastian Borgeaud, Trevor Cai +77
Language modelling provides a step towards intelligent communication systems by harnessing large repositories of written human knowledge to better predict and understand the world.…
Ethical and social risks of harm from Language Models
Laura Weidinger, John Mellor, Maribeth Rauh +20
This paper aims to help structure the risk landscape associated with large-scale Language Models (LMs). In order to foster advances in responsible innovation, an in-depth understan…
Improving language models by retrieving from trillions of tokens
Sebastian Borgeaud, Arthur Mensch, Jordan Hoffmann +25
We enhance auto-regressive language models by conditioning on document chunks retrieved from a large corpus, based on local similarity with preceding tokens. With a trillion to…
Alignment of Language Agents
Zachary Kenton, Tom Everitt, Laura Weidinger +3
For artificial intelligence to be beneficial to humans the behaviour of AI agents needs to be aligned with what humans want. In this paper we discuss some behavioural issues for la…
Fine-Tuning Language Models from Human Preferences
Daniel M. Ziegler, Nisan Stiennon, Jeffrey Wu +5
Reward learning enables the application of reinforcement learning (RL) to tasks where reward is defined by human judgment, building a model of reward by asking humans questions. Mo…
Reward learning from human preferences and demonstrations in Atari
Borja Ibarz, Jan Leike, Tobias Pohlen +3
To solve complex real-world problems with reinforcement learning, we cannot rely on manually specified reward functions. Instead, we can have humans communicate an objective to the…
AI safety via debate
Geoffrey Irving, Paul Christiano, Dario Amodei
To make AI systems broadly useful for challenging real-world tasks, we need them to learn complex human goals and preferences. One approach to specifying complex goals asks humans…
Deep Network Guided Proof Search
Sarah Loos, Geoffrey Irving, Christian Szegedy +1
Deep learning techniques lie at the heart of several significant AI advances in recent years including object recognition and detection, image captioning, machine translation, spee…
DeepMath - Deep Sequence Models for Premise Selection
Alex A. Alemi, Francois Chollet, Niklas Een +3
We study the effectiveness of neural sequence models for premise selection in automated theorem proving, one of the main bottlenecks in the formalization of mathematics. We propose…
TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
Martín Abadi, Ashish Agarwal, Paul Barham +37
TensorFlow is an interface for expressing machine learning algorithms, and an implementation for executing such algorithms. A computation expressed using TensorFlow can be executed…