Benjamin Lahner

Benjamin Lahner headshot

Education:

PhD Candidate: Electrical Engineering and Computer Science, MIT, Expected May 2025

Thesis: Understanding Human Visual Intelligence Through Large-scale Brain Imaging and Computational Models.

Advisor: Aude Oliva, MIT Computer Science and Artificial Intelligence Lab (CSAIL)


MS: Electrical Engineering and Computer Science, MIT, May 2022

BS: Biomedical Engineering, Boston University, May 2019

GitHub LinkedIn Google Scholar

Contact: blahner [at] mit [dot] edu

Current Role: PhD Candidate in MIT CSAIL for Computational Neuroscience. Expected to graduate in May 2025.

Research Summary:

My research combines neuroscience and AI to understand how our visual system recognizes actions, understands complex scenes, and remembers images. I've successfully applied my research in industry settings, including developing personalized conversational AI systems at Amazon, training large vision-language models at Microsoft, and predicting disease severity from patient data at Regeneron. Starting summer 2025, I aim to join an industry research team where I can combine my expertise in human and artificial intelligence to develop next-generation AI systems.

TreeCV: Click on a filled node (●) to expand/collapse children nodes. If node is clickable (🔗), click on the text to visit the URL.

Legend:

● Skills
● Work Experience
● Publications (*First Author)
● Patents
● Awards
Work Experience
Amazon Design Technologist Intern Winter 2024, Summer 2024
  • Built a product-focused conversational AI system by combining agentic teams of large language models (from OpenAI, Anthropic, and open source) with personalizable knowledge graphs and vector databases.
  • Improved the previous system's multi-hop retrieval recall score by 15% and temporal reasoning accuracy by 10% on large-scale benchmark datasets.
  • Prioritized a flexible, model-agnostic framework to evaluate and incoporate next-generation LLMs.
Microsoft Computer Vision Research Intern Summer 2022
  • Trained large transformer models to unite vision and language tasks.
  • Integrated in a future deployment to Microsoft's Azure AI platform.
Regeneron Pharmaceuticals Machine Learning PhD Intern Summer 2021
  • Developed machine learning models to predict clinically relevant outcomes (e.g., pain) from wearable sensor data.
  • Presented results to senior management, which heavily influenced critical investment decisions in the emerging field of wearable healthcare technologies.
  • Work resulted in being a co-inventor on a patent, "Systems and methods for test device analysis".
Publications
2024
Modeling short visual events through the BOLD moments video fMRI dataset and metadata.
Benjamin Lahner, Kshitij Dwivedi, Polina Iamshchinina, Monika Graumann, Alex Lascelles, Gemma Roig, Alessandro Thomas Gifford, Bowen Pan, SouYoung Jin, Ratan Murty, Kendrick Kay, Aude Oliva+, Radoslaw Cichy+.
Nature Communications, 2024
Paper | Project page | Dataset Repository
Brain Netflix: Scaling Data to Reconstruct Videos from Brain Signals.
Camilo Fosco*, Benjamin Lahner*, Bowen Pan, Alex Andonian, Emilie Josephs, Alex Lascelles, and Aude Oliva.
European Conference on Computer Vision (ECCV), 2024
Paper | Project page
Visual perception of highly memorable images is mediated by a distributed network of ventral visual regions that enable a late memorability response.
Benjamin Lahner*, Yalda Mohsenzadeh*, Caitlin Mullin*, and Aude Oliva.
PLOS Biology, 2024
Paper | Project page | Repository | MIT news
Digital wearable insole-based identification of knee arthropathies and gait signatures using machine learning.
Matthew F. Wipperman*, Allen Z. Lin*, Kaitlyn M. Gayvert*, Benjamin Lahner, ... , and Olivier Harari.
eLife, 2024
Paper
A mechanical device for precise self-administration of ocular drugs.
Jesse George-Akpenyi*, Benjamin Lahner*, Seung Hyeon Shim*, Carly Smith*, Nakul Singh, Matt Murphy, Leroy Sibanda, Giovanni Traverso, and Nevan C. Hanumara.
Human Factors in Healthcare, 2024
Paper | Repository | Patent App
2023
Theta-phase-specific modulation of dentate gyrus memory neurons.
Bahar Rahsepar, Jad Noueihed, Jacob F. Norman, Benjamin Lahner, Melanie H. Quick, Kevin Ghaemi, Aashna Pandya, Fernando R. Fernandez, Steve Ramirez, and John A. White.
eLife, 2023
Paper
The Algonauts Project 2023 Challenge: How the Human Brain Makes Sense of Natural Scenes.
Allesandro T. Gifford, Benjamin Lahner, Sari Saba-Sadiya, Martina G. Vilas, Alex Lascelles, Aude Oliva, Kendrick Kay, Gemma Roig, Radoslaw M. Cichy.
arXiv, 2023
Paper | Project page
2022
Cochlea to categories: The spatiotemporal dynamics of semantic auditory representations.
Matthew X. Lowe*, Yalda Mohsenzadeh*, Benjamin Lahner, Ian Charest, Aude Oliva and Santani Teng.
Cognitive Neuropsychology, 2022
Paper | Project page | Video
2021
The Algonauts Project 2021 Challenge: How the Human Brain Makes Sense of a World in Motion.
Radoslaw Martin Cichy, Kshitij Dwivedi, Benjamin Lahner, Alex Lascelles, Polina Iamshchinina, M Graumann, Alex Andonian, NAR Murty, K Kay, Gemma Roig, Aude Oliva.
arXiv, 2021
Paper | Project page
2020
Emergence of visual center-periphery spatial organization in deep convolutional neural networks.
Yalda Mohsenzadeh, Caitlin Mullin, Benjamin Lahner, and Aude Oliva.
Scientific Reports, 2020
Paper | Project page
2019
Reliability and generalizability of similarity-based fusion of fMRI and MEG data in the ventral and dorsal visual streams.
Yalda Mohsenzadeh*, Caitlin Mullin*, Benjamin Lahner, Radoslaw Cichy, and Aude Oliva.
Vision, 2019
Paper | Project page | Video
The Algonauts Project: A platform for communication between the sciences of biological and artificial intelligence.
Radoslaw Martin Cichy, Gemma Roig, Alex Andonian, Kshitij Dwivedi, Benjamin Lahner, Alex Lascelles, Yalda Mohsenzadeh, Kandan Ramakrishnan, and Aude Oliva.
arXiv, 2019
Paper | Project page
Patents
Eye drop positioning device with haptic feedback
May 2023
  • Designed a mechanical eye drop assist device for elderly glaucoma patients.
  • Published results in the journal Human Factors in Healthcare.
  • International Patent Application: WO/2024/238546
Systems and methods for test device analysis
August 2023
  • Created a framework for testing machine learning models used to predict clinical outcomes from wearable devices.
  • Worked in collaboration with Regeneron Pharmaceuticals and published results in the journal eLife.
  • Non-provisional application in progress.
Awards
MIT Open Data Competition - Runner Up
Fall 2022
  • Runner up (out of 70 projects across MIT) in the Open Data competition that recognizes open and publicly accessible data with strong potential for large scientific impact.
EECS Mathworks Fellowship
Fall 2022
  • Awarded full financial support for one academic year ($70,000) for using MATLAB in novel and impactful scientific research.
Best Biomedical Engineering Senior Design Project
Spring 2019
  • Developed and deployed a real-time algorithm (latency of ~20ms) in C++ that interfaced with neural signals from a mouse's hippocampus to manipulate memory encoding and retrieval. Work published in the journal eLife.
  • Awarded best project out of 42 other projects in biomedical engineering by engineering faculty.