Diane Tchuindjo

Ph.D. Student at MIT

Research

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.

Contact

Email me at 'd <full-last-name>' at gmail dot com

Random Musings

For more of my thoughts, check out the musings page.

Papers

• OBLIQ-Bench: Exposing Overlooked Bottlenecks in Modern Retrievers with Latent and Implicit Queries
  by Diane Tchuindjo, Devavrat Shah, and Omar Khattab
  Retrieval benchmarks are increasingly saturating, but we argue that efficient search is far from a solved problem. We identify a class of queries we call oblique, which seek documents that instantiate a latent pattern, like finding all tweets that express an implicit stance, chat logs that demonstrate a particular failure mode, or transcripts that match an abstract scenario. We study three mechanisms through which obliqueness may arise and introduce OBLIQ-Bench, a suite of five oblique search problems over real long-tail corpora.
  Paper. Benchmark.
• Reasoning-Intensive Regression
  by Diane Tchuindjo and Omar Khattab
  We introduce Reasoning-Intensive Regression (RiR), regression tasks requiring precise numerical predictions that depend on multi-step reasoning. LLMs reason well but output coarsely; small regressors calibrate well but can't reason. We develop methods to bridge this gap.
  Blog post. Paper. NeurIPS Efficient Reasoning Workshop '25, ACM CAIS '26 (Industry Spotlight).