Aarush Sinha

$ query := "information retrieval"
retrieved 0 spans · type a query or click a tag

about

Aarush Sinha

I am a Master's student in Computer Science at the University of Copenhagen, specializing in Information Retrieval and Natural Language Processing, where I am advised by Prof. Maria Maistro and Prof. Christina Lioma. I previously completed my undergraduate at VIT Chennai.

My research is centered around the retrieval and generative capabilites of models particularly focusing on its shortcomings and efficiency. I am currently interning at LightOn, where I work on code search tools.

Over the years, I have had the opportunity to collaborate with wonderful mentors across several labs. I previously worked on hallucinations in Text2Video models with Prof. Amitava Das at the AI Institute, UofSC, and on efficient dense retrievers with Prof. Nirav Bhatt at WSAI-IIT Madras. Most recently, at the Stanford STAIR Lab, I worked with Rylan Schaeffer and Prof. Sanmi Koyejo on model collapse.

Research Interests

Service

SRW@ACL'25 (Reviewer) FrontierIR@AAAI'26 (Reviewer) COLM'26 (Reviewer) SIGIR'26 (Reviewer)

updates

June 2026
Interning at LightOn, working on code search tools.
June 2026
New pre-print released on arXiv: Mechanistic Analysis of Alignment Algorithms in Language Models.
June 2026
New pre-print released on arXiv: Evaluation Cards: An Interpretive Layer for AI Evaluation Reporting.
March 2026
New pre-print released on arXiv: ECIsem: Semantic Residual Effective Contrastive Information for Evaluating Hard Negatives.