publications

2026

1. ICML

   A Benchmark and Framework for Evaluating Next Action Predictions in Spreadsheets

   Tejas Agrawal, Vu Le, Sumit Gulwani, and 1 more author

   In International Conference on Machine Learning (ICML) , 2026

   Abs arXiv PDF Code

   We introduce a benchmark and online evaluation framework for predicting users’ next actions in spreadsheets. The benchmark contains 52 curated editing sequences comprising 12K actions. We compare LLMs, fine-tuned small language models, and classical predictors, studying action savings, false positives, efficiency, user profiles, prediction triggers, and context.

2024

1. 

   Grasping Graphormer : Assessing Transformer Performance for Graph Representation

   Tejas Agrawal, Karan Bania, Yash Bhisikar, and 3 more authors

   Jun 2024

   Abs HTML

   A first-principles blog post to understand the Graphormer.

2. 

   CountCLIP – [Re] Teaching CLIP to Count to Ten

   Harshvardhan Mestha, Tejas Agrawal, Karan Bania, and 2 more authors

   Jun 2024

   Abs HTML Code

   Large vision-language models (VLMs) are shown to learn rich joint image-text representations enabling high performances in relevant downstream tasks. However, they fail to showcase their quantitative understanding of objects, and they lack good counting-aware representation. This paper conducts a reproducibility study of ’Teaching CLIP to Count to Ten’ (Paiss et al., 2023), which presents a method to finetune a CLIP model (Radford et al., 2021) to improve zero-shot counting accuracy in an image while maintaining the performance for zero-shot classification by introducing a counting-contrastive loss term. We improve the model’s performance on a smaller subset of their training data with lower computational resources. We verify these claims by reproducing their study with our own code. The implementation can be found at [this URL](https://github.com/SforAiDl/CountCLIP).