Diane Tchuindjo

Figuring things out slowly.

I'm currently a PhD student at MIT, advised by Omar Khattab and Devavrat Shah.

My research focuses on designing algorithms for reasoning-intensive prediction in NLP, especially systems that transform latent reasoning into reliable scores, rankings, and retrieval decisions.

Papers

• OBLIQ-Bench: Exposing Overlooked Bottlenecks in Modern Retrievers with Latent and Implicit Queries
  Tchuindjo D., Shah D., Khattab O.
  Paper. Benchmark. Blog post.
• Reasoning-Intensive Regression
  Tchuindjo D., Khattab O.
  Paper. Blog post. ACM CAIS '26 (Industry Spotlight).

Musings

• Let's Not Be ObliqueAugust 17, 2026
• The Curious Case of LexJanuary 17, 2026
• Let's Reason IntensivelyDecember 31, 2025
• A Law of RobustnessDecember 12, 2024
• Kleesia: A Hacky InterpretationOctober 29, 2024
• Honey, I Shrunk the..July 28, 2024
• A bit of ZigMay 19, 2024
• Anonymity is MarvelousApril 1, 2024
• How to Trust Yourself, EnoughFebruary 8, 2024
• Debugging CAS SoftwareJanuary 5, 2024
• I Don't Like Hard ProblemsJanuary 1, 2024