arXiv subjects

Sanket Mhatre

Publications and source records attributed to Sanket Mhatre.

3 recordsLinked to original sources

arXiv2026

fxBench: Evaluating and Understanding Formula Suggestions in Spreadsheets

Spreadsheets are among the most widely used tools for computations, and formulas are their core building blocks. Besides agentic spreadsheet tools, proactive formula suggestions enable significant productivity gains and are now available in popular spreadsheet software (Microsoft Excel and Google Sheets). Such suggestions pose interesting challenges: the right context must be collected and represented for a machine learning model, which must then be able to write a correct formula, in a way that is fast and cheap enough to trigger on every = (or subsequent keystrokes for auto-completion) without taking users out of their editing flow. Surprisingly, there is no public benchmark for formula suggestions on real spreadsheets. In this paper, we therefore (1) introduce fxBench as a manually curated benchmark of 503 formula suggestion tasks based on Sheetpedia, (2) break down the formula suggestion problem into four capabilities (gather context, understand context, understand intent, write formula) that a formula suggestion system requires, and (3) evaluate existing and new approaches to understand bottlenecks in these capabilities that pose interesting research directions. Specifically, we annotate each benchmark with the cells that describe intent to determine the theoretical performance limit of a given context selection strategy, we evaluate performance of small and large models on different representations of context, including a dense format with high theoretical coverage at low token counts, and we fine-tune smaller language models to show that we can learn to understand such dense formats. Additionally, we perform quantitative and qualitative analysis on failure modes to understand bottlenecks in the formula writing capability; the best configuration only achieved 58.8% correct suggestions.

cs.LG↗

arXiv2025

UnWEIRDing LLM Entity Recommendations

Large Language Models have been widely been adopted by users for writing tasks such as sentence completions. While this can improve writing efficiency, prior research shows that LLM-generated suggestions may exhibit cultural biases which may be difficult for users to detect, especially in educational contexts for non-native English speakers. While such prior work has studied the biases in LLM moral value alignment, we aim to investigate cultural biases in LLM recommendations for real-world entities. To do so, we use the WEIRD (Western, Educated, Industrialized, Rich and Democratic) framework to evaluate recommendations by various LLMs across a dataset of fine-grained entities, and apply pluralistic prompt-based strategies to mitigate these biases. Our results indicate that while such prompting strategies do reduce such biases, this reduction is not consistent across different models, and recommendations for some types of entities are more biased than others.

cs.CY↗

arXiv2025

SWE-Sharp-Bench: A Reproducible Benchmark for C# Software Engineering Tasks

AI coding agents have shown great progress on Python software engineering benchmarks like SWE-Bench, and for other languages like Java and C in benchmarks like Multi-SWE-Bench. However, C# -- a prominent enterprise language ranking #5 in the TIOBE index -- remains absent from such benchmarks. We introduce SWE-Sharp-Bench, a reproducible software engineering benchmark for C# featuring 150 instances from 17 repositories. Evaluating identical model-agent configurations across languages reveals a significant performance gap: while 70% of Python tasks in SWE-Bench Verified are solved, only 40% of our C# tasks are resolved. We open-source SWE-Sharp-Bench and our entire curation pipeline.

cs.SE↗