Natural Language Processing
Language models have become remarkably fluent. Fluency, however, is not the same as understanding. Our natural language research examines how meaning is represented computationally, how it connects to the world, and how language can carry reasoning rather than merely resemble it.
- Language understanding
- Computational semantics
- Multilingual modeling
- Grounded language learning
- Information retrieval
- Language reasoning
- Speech–language systems
Why it matters
Language is how people state goals, share knowledge, and justify decisions. Systems that work with language therefore sit close to human intent — which makes their failure modes consequential. A model that misreads an instruction, invents a citation, or reasons fluently to a false conclusion can do so convincingly.
There is also a question of reach. Most language technology performs best in a handful of high-resource languages. Building models that represent meaning in ways that transfer across languages is both a scientific problem and a matter of who benefits from the technology.
Key research questions
What evidence would distinguish a model that represents meaning from one that models surface co-occurrence?
How can retrieval and generation be combined so that claims remain traceable to sources?
Do multilingual models learn shared semantic representations, and how can low-resource languages benefit from them?
When a model produces a step-by-step explanation, how faithfully does it reflect the computation that produced the answer?
How should language be grounded in perception and action to support reference to the physical world?
Primary subfields
- Computational semantics
- Formal and learned accounts of meaning, entailment, and compositionality.
- Multilingual modeling
- Shared representations across languages and transfer to under-resourced ones.
- Retrieval and grounding
- Connecting language to documents, databases, and perceptual input.
- Reasoning in language
- Multi-step inference, argumentation, and the faithfulness of explanations.
- Speech and language
- Joint modeling of spoken and written language.
Representative methodologies
Targeted evaluation sets
Minimal pairs and controlled variations that test one linguistic phenomenon at a time.
Attention and attribution analysis
Tracing which inputs a model relies on for a given output.
Retrieval-augmented systems
Coupling models to external sources and measuring attribution quality.
Cross-lingual transfer
Training on some languages and evaluating on others.
Relationships to other domains
- Generative AI
Large language models are the shared platform for both fields.
- Computer Vision
Vision–language models ground words in visual evidence.
- Artificial Intelligence
Language is increasingly the medium in which agents plan and communicate.
Working on questions in Natural Language Processing?
We welcome conversations with researchers and organizations working on related questions — whether to compare notes, discuss a collaboration, or share work in progress.