Publications (27)
Character-level Chinese Backpack Language Models
Hao Sun, John Hewitt
The Backpack is a Transformer alternative shown to improve interpretability in English language modeling by decomposing predictions into a weighted sum of token sense components. H…
Probing artificial neural networks: insights from neuroscience
Anna A. Ivanova, John Hewitt, Noga Zaslavsky
A major challenge in both neuroscience and machine learning is the development of useful tools for understanding complex information processing systems. One such tool is probes, i.…
New Identification of the Mixed-Morphology Supernova Remnant G298.6-0.0 with Possible Gamma-ray Association
Aya Bamba, Makoto Sawada, Yuto Nakano +4
We present an X-ray analysis on the Galactic supernova remnant (SNR) G298.6-0.0 with Suzaku. The X-ray image shows a center-filled structure inside the radio shell, implying this S…
Lost in the Middle: How Language Models Use Long Contexts
Nelson F. Liu, Kevin Lin, John Hewitt +4
While recent language models have the ability to take long contexts as input, relatively little is known about how well they use longer context. We analyze the performance of langu…
Conditional probing: measuring usable information beyond a baseline
John Hewitt, Kawin Ethayarajh, Percy Liang +1
Probing experiments investigate the extent to which neural representations make properties -- like part-of-speech -- predictable. One suggests that a representation encodes a prope…
Improving Parametric Knowledge Access in Reasoning Language Models
Melody Ma, John Hewitt
We study reasoning for accessing world knowledge stored in a language model's parameters. For example, recalling that Canberra is Australia's capital may benefit from thinking thro…
Neologism Learning for Controllability and Self-Verbalization
John Hewitt, Oyvind Tafjord, Robert Geirhos +1
Humans invent new words when there is a rising demand for a new useful concept (e.g., doomscrolling). We explore and validate a similar idea in our communication with LLMs: introdu…
Finding Universal Grammatical Relations in Multilingual BERT
Ethan A. Chi, John Hewitt, Christopher D. Manning
Recent work has found evidence that Multilingual BERT (mBERT), a transformer-based multilingual masked language model, is capable of zero-shot cross-lingual transfer, suggesting th…
The EOS Decision and Length Extrapolation
Benjamin Newman, John Hewitt, Percy Liang +1
Extrapolation to unseen sequence lengths is a challenge for neural generative models of language. In this work, we characterize the effect on length extrapolation of a modeling dec…
Designing and Interpreting Probes with Control Tasks
John Hewitt, Percy Liang
Probes, supervised models trained to predict properties (like parts-of-speech) from representations (like ELMo), have achieved high accuracy on a range of linguistic tasks. But doe…
XNMT: The eXtensible Neural Machine Translation Toolkit
Graham Neubig, Matthias Sperber, Xinyi Wang +10
This paper describes XNMT, the eXtensible Neural Machine Translation toolkit. XNMT distin- guishes itself from other open-source NMT toolkits by its focus on modular code design, w…
Subliminal Steering: Stronger Encoding of Hidden Signals
George Morgulis, John Hewitt
Subliminal learning describes a student language model inheriting a behavioral bias by fine-tuning on seemingly innocuous data generated by a biased teacher model. Prior work has b…
Simple, Fast, Accurate Intent Classification and Slot Labeling for Goal-Oriented Dialogue Systems
Arshit Gupta, John Hewitt, Katrin Kirchhoff
With the advent of conversational assistants, like Amazon Alexa, Google Now, etc., dialogue systems are gaining a lot of traction, especially in industrial setting. These systems t…
Backpack Language Models
John Hewitt, John Thickstun, Christopher D. Manning +1
We present Backpacks: a new neural architecture that marries strong modeling performance with an interface for interpretability and control. Backpacks learn multiple non-contextual…
Instruction Following without Instruction Tuning
John Hewitt, Nelson F. Liu, Percy Liang +1
Instruction tuning commonly means finetuning a language model on instruction-response pairs. We discover two forms of adaptation (tuning) that are deficient compared to instruction…
Learning Translations via Matrix Completion
Derry Wijaya, Brendan Callahan, John Hewitt +4
Bilingual Lexicon Induction is the task of learning word translations without bilingual parallel corpora. We model this task as a matrix completion problem, and present an effectiv…
On the Opportunities and Risks of Foundation Models
Rishi Bommasani, Drew A. Hudson, Ehsan Adeli +111
AI is undergoing a paradigm shift with the rise of models (e.g., BERT, DALL-E, GPT-3) that are trained on broad data at scale and are adaptable to a wide range of downstream tasks.…
JamPatoisNLI: A Jamaican Patois Natural Language Inference Dataset
Ruth-Ann Armstrong, John Hewitt, Christopher Manning
JamPatoisNLI provides the first dataset for natural language inference in a creole language, Jamaican Patois. Many of the most-spoken low-resource languages are creoles. These lang…
Refining Targeted Syntactic Evaluation of Language Models
Benjamin Newman, Kai-Siang Ang, Julia Gong +1
Targeted syntactic evaluation of subject-verb number agreement in English (TSE) evaluates language models' syntactic knowledge using hand-crafted minimal pairs of sentences that di…
Truncation Sampling as Language Model Desmoothing
John Hewitt, Christopher D. Manning, Percy Liang
Long samples of text from neural language models can be of poor quality. Truncation sampling algorithms--like top- or top- -- address this by setting some words' probabilitie…
Model Editing with Canonical Examples
John Hewitt, Sarah Chen, Lanruo Lora Xie +3
We introduce model editing with canonical examples, a setting in which (1) a single learning example is provided per desired behavior, (2) evaluation is performed exclusively out-o…
RNNs can generate bounded hierarchical languages with optimal memory
John Hewitt, Michael Hahn, Surya Ganguli +2
Recurrent neural networks empirically generate natural language with high syntactic fidelity. However, their success is not well-understood theoretically. We provide theoretical in…
The Third Fermi Large Area Telescope Catalog of Gamma-ray Pulsars
David A. Smith, Philippe Bruel, Colin J. Clark +156
We present 294 pulsars found in GeV data from the Large Area Telescope (LAT) on the Fermi Gamma-ray Space Telescope. Another 33 millisecond pulsars (MSPs) discovered in deep radio…
Closing the Curious Case of Neural Text Degeneration
Matthew Finlayson, John Hewitt, Alexander Koller +2
Despite their ubiquity in language generation, it remains unknown why truncation sampling heuristics like nucleus sampling are so effective. We provide a theoretical explanation fo…
Because we have LLMs, we Can and Should Pursue Agentic Interpretability
Been Kim, John Hewitt, Neel Nanda +2
The era of Large Language Models (LLMs) presents a new opportunity for interpretability--agentic interpretability: a multi-turn conversation with an LLM wherein the LLM proactively…
We Can't Understand AI Using our Existing Vocabulary
John Hewitt, Robert Geirhos, Been Kim
This position paper argues that, in order to understand AI, we cannot rely on our existing vocabulary of human words. Instead, we should strive to develop neologisms: new words tha…
Discovery of X-ray Emission from the Galactic Supernova Remnant G32.8-0.1 with Suzaku
Aya Bamba, Yukikatsu Terada, John Hewitt +6
We present the first dedicated X-ray study of the supernova remnant (SNR) G32.8-0.1 (Kes 78) with Suzaku. X-ray emission from the whole SNR shell has been detected for the first ti…