Sotabase

Career

· Research Scientist, Anthropic2023–
· Research Intern, DeepMind2022–
· Research Intern, FAIR, Meta2021–
· Visiting Researcher, MIT LINGO2020–
· Course Consultant, Codecademy2019–2020
· PhD in Computer Science, Stanford University2018–2023
· Intern, Amazon Alexa AI2017–
· MPhil in Advanced Computer Science, University of Cambridge2017–2018
· B.A. in Computer Science, Boston College2013–2017
· Anthropic

Publications (44)

Learning to Compress Prompts with Gist Tokens
Neural Information Processing Systems · 2023
304
cited
Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
arXiv.org · 2024
285
cited
Many-shot Jailbreaking
Neural Information Processing Systems · 2024
214
cited
Compositional Explanations of Neurons
Neural Information Processing Systems · 2020
204
cited
Parkinson's Disease Subtypes Identified from Cluster Analysis of Motor and Non-motor Symptoms
Frontiers in Aging Neuroscience · 2017
130
cited
Improving Intrinsic Exploration with Language Abstractions
Neural Information Processing Systems · 2022
73
cited
Emergent Communication of Generalizations
Neural Information Processing Systems · 2021
50
cited
Shaping Visual Representations with Language for Few-Shot Classification
Annual Meeting of the Association for Computational Linguistics · 2019
49
cited
Active Learning Helps Pretrained Models Learn the Intended Task
Neural Information Processing Systems · 2022
46
cited
Learning to refer informatively by amortizing pragmatic reasoning
Annual Meeting of the Cognitive Science Society · 2020
24
cited
The Meta-Science of Adult Statistical Word Segmentation: Part 1
Collabra: Psychology · 2019
21
cited
UvA-DARE (Digital Academic Repository) Learning outside the box: Discourse-level features improve metaphor identification
17
cited
Improving Policy Learning via Language Dynamics Distillation
Neural Information Processing Systems · 2022
16
cited
Do We Need Natural Language?: Exploring Restricted Language Interfaces for Complex Domains
CHI Extended Abstracts · 2019
15
cited
Calibrate your listeners! Robust communication-based training for pragmatic speakers
Conference on Empirical Methods in Natural Language Processing · 2021
7
cited
Jailbreak Defense in a Narrow Domain: Limitations of Existing Methods and a New Transcript-Classifier Approach
arXiv.org · 2024
6
cited
I N THE ZONE: M EASURING DIFFICULTY AND PROGRESSION IN CURRICULUM GENERATION
2022
3
cited
Characterizing tradeoffs between teaching via language and demonstrations in multi-agent systems
Annual Meeting of the Cognitive Science Society · 2023
2
cited
Evaluating Hierarchies of Verb Argument Structure with Hierarchical Clustering
Conference on Empirical Methods in Natural Language Processing · 2017
1
cited
Replication of Frank, Goldwater, Griffiths, & Tenenbaum (2010): Modeling human performance in statistical word segmentation, Experiment 1
2017
1
cited
Sotabase
Jesse Mu | Researcher Profile | Sotabase | Sotabase