ELISA KREISS
ekreiss@stanford.edu
A portrait of Elisa, a white woman with blond hair and thin gold glasses, smiling into the camera. She is shown from the elbows up wearing a white blouse worn openly on top of a black top with a thin three-layered golden necklace and thin hoop earrings.She is standing in sunny Cupertino. The background is blurred.

I'm a 5th year Linguistics PhD student at Stanford where I investigate how we produce and understand language situated in the visual world. My work combines tools from natural language processing, psycholinguistics, and human-computer interaction to advance our understanding of how communicative context shapes language use. My research has direct applications to image accessibility – the challenge of (automatically) generating image descriptions for blind and low vision users.

I'm a member of Stanford's NLP group (advised by Christopher Potts) and I'm further affiliated with the CoCoLab, ALPS lab, and Causality in Cognition Lab.

Find out more: CV, Google Scholar, Twitter, Github.

Publications

Context matters for image descriptions for accessibility: Challenges for referenceless evaluation metrics
Elisa Kreiss, Cynthia Bennett, Shayan Hooshmand, Eric Zelikman, Meredith Ringel Morris, and Christopher Potts (2022)
To appear at Empirical Methods in Natural Language Processing (EMNLP)
[ Paper | Repository ]
Concadia: Towards image-based text generation with a purpose
Elisa Kreiss, Fei Fang, Noah D. Goodman, and Christopher Potts (2022)
To appear at Empirical Methods in Natural Language Processing (EMNLP)
[ Paper | Repository | Corpus Exploration ]
Color overmodification emerges from data-driven learning and pragmatic reasoning
Fei Fang, Kunal Sinha, Noah D. Goodman, Christopher Potts, Elisa Kreiss (2022)
Proceedings of the 44th Annual Conference of the Cognitive Science Society (CogSci)
[ Paper | Repository ]
Causal distillation for language models
Zhengxuan Wu*, Atticus Geiger*, Josh Rozner, Elisa Kreiss, Hanson Lu, Thomas Icard, Christopher Potts, and Noah D. Goodman (2022)
North American Chapter of the Association for Computational Linguistics (NAACL)
[ Paper | Repository ]
Inducing causal structure for interpretable neural networks
Atticus Geiger*, Zhengxuan Wu*, Hanson Lu*, Josh Rozner, Elisa Kreiss, Thomas Icard, Noah D. Goodman, and Christopher Potts (2022)
International Conference on Machine Learning (ICML)
[ Paper | Repository ]
ReaSCAN: Compositional reasoning in language grounding
Zhengxuan Wu*, Elisa Kreiss*, Desmond Ong, and Christopher Potts (2021)
NeurIPS Track on Datasets and Benchmarks
[ Paper | Repository | Dataset Website ]
Intuitive image descriptions are context-sensitive
Shayan Hooshmand, Elisa Kreiss, and Christopher Potts (2021)
NeurIPS Workshop: Meaning in Context
[ Extended Abstract ]
Modeling subjective assessments of guilt in newspaper crime narratives
Elisa Kreiss*, Zijian Wang*, and Christopher Potts (2020)
Proceedings of the 24th Conference on Computational Natural Language Learning (CoNLL)
[ Paper | Repository ]
Production expectations modulate contrastive inference
Elisa Kreiss, and Judith Degen (2020)
Proceedings of the 42nd Annual Conference of the Cognitive Science Society (CogSci)
[ Paper | Repository | Explore the Data | Talk ]
When redundancy is useful: A Bayesian approach to "overinformative" referring expressions
Judith Degen, Robert D. Hawkins, Caroline Graf, Elisa Kreiss, and Noah D. Goodman (2020)
Psychological Review
[ Paper (open access) | Repository ]
Uncertain evidence statements and guilt perception in iterative reproductions of crime stories
Elisa Kreiss, Michael Franke, and Judith Degen (2019)
Proceedings of the 41st Annual Conference of the Cognitive Science Society (CogSci)
[ Paper | Repository ]

© Elisa Kreiss