242 citations · 1.5k across the 60 of their papers we have counts for
38 papers · 1 filter
FIND: A Function Description Benchmark for Evaluating Interpretability Methods
Sarah Schwettmann, Tamar Rott Shaham, Joanna Materzynska +5
Labeling neural network submodules with human-legible descriptions is useful for many downstream tasks: such descriptions can surface failures, guide interventions, and perhaps eve…
Linearity of Relation Decoding in Transformer Language Models
Evan Hernandez, Arnab Sen Sharma, Tal Haklay +5
Much of the knowledge encoded in transformer language models (LMs) may be expressed in terms of relations: relations between words and their synonyms, entities and their attributes…
Reasoning or Reciting? Exploring the Capabilities and Limitations of Language Models Through Counterfactual Tasks
Zhaofeng Wu, Linlu Qiu, Alexis Ross +6
The impressive performance of recent language models across a wide range of tasks suggests that they possess a degree of abstract reasoning skills. Are these skills general and tra…
From Word Models to World Models: Translating from Natural Language to the Probabilistic Language of Thought
Lionel Wong, Gabriel Grand, Alexander K. Lew +4
How does language inform our downstream thinking? In particular, how do humans make meaning from language--and how can we leverage a theory of linguistic meaning to build machines…
Decision-Oriented Dialogue for Human-AI Collaboration
Jessy Lin, Nicholas Tomlin, Jacob Andreas +1
We describe a class of tasks called decision-oriented dialogues, in which AI assistants such as large language models (LMs) must collaborate with one or more humans via natural lan…
Natural Language Decomposition and Interpretation of Complex Utterances
Harsh Jhamtani, Hao Fang, Patrick Xia +3
Designing natural language interfaces has historically required collecting supervised data to translate user requests into carefully designed intent representations. This requires…
Inspecting and Editing Knowledge Representations in Language Models
Evan Hernandez, Belinda Z. Li, Jacob Andreas
Neural language models (LMs) represent facts about the world described by text. Sometimes these facts derive from training data (in most LMs, a representation of the word "banana"…
Language Models Trained on Media Diets Can Predict Public Opinion
Eric Chu, Jacob Andreas, Stephen Ansolabehere +1
Public opinion reflects and shapes societal behavior, but the traditional survey-based tools to measure it are limited. We introduce a novel approach to probe media diet models --…
Language Modeling with Latent Situations
Belinda Z. Li, Maxwell Nye, Jacob Andreas
Language models (LMs) often generate incoherent outputs: they refer to events and entity states that are incompatible with the state of the world described in their inputs. We intr…
Language Models as Agent Models
Jacob Andreas
Language models (LMs) are trained on collections of documents, written by individual human agents to achieve specific goals in an outside world. During training, LMs have access on…
PromptBoosting: Black-Box Text Classification with Ten Forward Passes
Bairu Hou, Joe O'Connor, Jacob Andreas +2
We describe PromptBoosting, a query-efficient procedure for building a text classifier from a neural language model (LM) without access to the LM's parameters, gradients, or hidden…
Hierarchical Phrase-based Sequence-to-Sequence Learning
Bailin Wang, Ivan Titov, Jacob Andreas +1
We describe a neural transducer that maintains the flexibility of standard sequence-to-sequence (seq2seq) models while incorporating hierarchical phrases as a source of inductive b…
The Whole Truth and Nothing But the Truth: Faithful and Controllable Dialogue Response Generation with Dataflow Transduction and Constrained Decoding
Hao Fang, Anusha Balakrishnan, Harsh Jhamtani +7
In a real-world dialogue system, generated text must be truthful and informative while remaining fluent and adhering to a prescribed style. Satisfying these constraints simultaneou…
Identifying concept libraries from language about object structure
Catherine Wong, William P. McCarthy, Gabriel Grand +5
Our understanding of the visual world goes beyond naming objects, encompassing our ability to parse objects into meaningful parts, attributes, and relations. In this work, we lever…
Towards Tracing Factual Knowledge in Language Models Back to the Training Data
Ekin Akyürek, Tolga Bolukbasi, Frederick Liu +4
Language models (LMs) have been shown to memorize a great deal of factual knowledge contained in their training data. But when an LM generates an assertion, it is often difficult t…
Compositionality as Lexical Symmetry
Ekin Akyürek, Jacob Andreas
In tasks like semantic parsing, instruction following, and question answering, standard deep networks fail to generalize compositionally from small datasets. Many existing approach…
Quantifying Adaptability in Pre-trained Language Models with 500 Tasks
Belinda Z. Li, Jane Yu, Madian Khabsa +3
When a neural language model (LM) is adapted to perform a new task, what aspects of the task predict the eventual performance of the model? In NLP, systematic features of LM genera…
FairyTailor: A Multimodal Generative Framework for Storytelling
Eden Bensaid, Mauro Martino, Benjamin Hoover +1
Storytelling is an open-ended task that entails creative thinking and requires a constant flow of ideas. Natural language generation (NLG) for storytelling is especially challengin…
What Context Features Can Transformer Language Models Use?
Joe O'Connor, Jacob Andreas
Transformer-based language models benefit from conditioning on contexts of hundreds to thousands of previous tokens. What aspects of these contexts contribute to accurate model pre…
Lexicon Learning for Few-Shot Neural Sequence Modeling
Ekin Akyürek, Jacob Andreas
Sequence-to-sequence transduction is the core problem in language processing applications as diverse as semantic parsing, machine translation, and instruction following. The neural…
Implicit Representations of Meaning in Neural Language Models
Belinda Z. Li, Maxwell Nye, Jacob Andreas
Does the effectiveness of neural language models derive entirely from accurate modeling of surface word co-occurrence statistics, or do these models represent and reason about the…
The Low-Dimensional Linear Geometry of Contextualized Word Representations
Evan Hernandez, Jacob Andreas
Black-box probing models can reliably extract linguistic features like tense, number, and syntactic role from pretrained word representations. However, the manner in which these fe…
Multitasking Inhibits Semantic Drift
Athul Paul Jacob, Mike Lewis, Jacob Andreas
When intelligent agents communicate to accomplish shared goals, how do these goals shape the agents' language? We study the dynamics of learning in latent language policies (LLPs),…
Learning to Recombine and Resample Data for Compositional Generalization
Ekin Akyürek, Afra Feyza Akyürek, Jacob Andreas
Flexible neural sequence models outperform grammar- and automaton-based counterparts on a variety of tasks. However, neural models perform poorly in settings requiring compositiona…