Siddharth Yayavaram

I'm currently pursuing a Master's in Natural Language Processing and Machine Learning at Carnegie Mellon University (GPA: 4.25/4). I earned my Bachelor's in Computer Science from BITS Pilani, graduating as the Institute Gold Medalist (Rank 1) with a GPA of 9.97/10.

My research spans multimodal AI, LLM agents, evaluation, and culturally aware machine learning systems. Recently, I worked on GameDevBench, a benchmark for evaluating agentic capabilities through game development, which was accepted to ICML 2026.

For my undergraduate thesis, I worked with Graham Neubig, Simran Khanuja, and Michael Saxon to build a system that augments vision-language models with information from a multilingual, multicultural knowledge base, improving performance on scoring images for cultural relevance. This work was accepted to CEGIS @ ICCV 2025 and EACL 2026 (main conference).

My prior research has explored machine learning applications in mental health using speech data (IEEE IS 2024), post-training language models for idiom classification (MWE-UD @ LREC-COLING 2024), and depression detection.

I am currently looking for full-time Machine Learning Engineering, ML Research, and NLP Research roles, starting December 2026. Feel free to reach out if you think I'd be a good fit!

syayavar@cs.cmu.edu  /  CV  /  LinkedIn  /  GitHub

Publications

* indicates equal contribution / co-first authorship

GameDevBench GameDevBench: Evaluating Agentic Capabilities Through Game Development
Wayne Chi, Yixiong Fang, Arnav Yayavaram, Siddharth Yayavaram, Seth Karten, Qiuhong Anna Wei, Runkun Chen, Alexander Wang, Valerie Chen, Ameet Talwalkar, Chris Donahue
ICML, 2026
[Paper]   [Code]
CAIRE CAIRE: Cultural Attribution of Images by Retrieval-Augmented Evaluation
*Arnav Yayavaram, *Siddharth Yayavaram, *Simran Khanuja, Michael Saxon, Graham Neubig
CEGIS @ ICCV, 2025  ·  EACL (Main Conference), 2026
[Paper]   [Code]
BERT Idiom BERT-based Idiom Identification using Language Translation and Word Cohesion
*Arnav Yayavaram, *Siddharth Yayavaram, Prajna Upadhyay, Apurba Das
MWE-UD @ LREC-COLING, 2024
[Paper]   [Code]
IEEE IS Interpretable Feature Optimization for Sadness Recognition in Speech Emotion Analysis
*Siddharth Yayavaram, *Arnav Yayavaram, Jabez Christopher, Vasan Arunachalam
IEEE Intelligent Systems, 2024
[Paper]   [Code]

Template from Jon Barron.