--- title: "Rahul K. Arora — author profile and papers" description: "Rahul K. Arora: 2 papers on Elamur Research (2025) — affiliations, topics, co-authors and who cites them in our library." canonical: "https://research.elamur.ai/free/en/authors/rahul-k-arora/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/rahul-k-arora/) [home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / Rahul K. Arora # Rahul K. Arora Rahul K. Arora's papers here concern large language models: how they are built and released, and how their behaviour is measured in settings where errors carry real costs. The work spans [language models and pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) and [reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/). It ranges from describing the training and safety choices behind openly released reasoning systems to designing evaluations that judge model answers against expert standards rather than fixed answer keys. The most cited of these papers is the [gpt-oss-120b & gpt-oss-20b Model Card](https://md.research.elamur.ai/free/en/papers/agarwal-2025-gpt-oss-120b-gpt-oss/). It documents two open-weight, text-only Mixture-of-Experts reasoning models from OpenAI that are aimed at agentic use such as browsing, running Python and calling functions. The card explains that users can adjust how much reasoning effort the models spend and that their chains of thought are exposed. It also describes how dangerous biosecurity content was filtered from pretraining and how post-training relied on chain-of-thought reinforcement learning. The second is [HealthBench](https://md.research.elamur.ai/free/en/papers/arora-2025-healthbench-evaluating-large-language-models/), a benchmark built from realistic health conversations with both lay users and clinicians, in many languages and personas. Each conversation is scored against its own physician-written rubric by a model-based grader, and that grader's verdicts are checked against physician judgment. Both works date from 2025 and point in the same direction: pairing more capable reasoning models with careful accounting of their risks and of their usefulness in health care. Co-authors on these papers include [Karan Singhal](https://md.research.elamur.ai/free/en/authors/karan-singhal/), [Johannes Heidecke](https://md.research.elamur.ai/free/en/authors/johannes-heidecke/), [Jason Wei](https://md.research.elamur.ai/free/en/authors/jason-wei/) and [Sam Altman](https://md.research.elamur.ai/free/en/authors/sam-altman/), all of whom appear alongside Arora on one paper each. Rahul K. Arora has 2 papers in our library, in 2025, most of them on [Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/), [Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/); often with [Alex Beutel](https://md.research.elamur.ai/free/en/authors/alex-beutel/), [Bowen Baker](https://md.research.elamur.ai/free/en/authors/bowen-baker/), [Jason Wei](https://md.research.elamur.ai/free/en/authors/jason-wei/). 2 papers in our library · 2025 No OpenAlex profile matched yet. [2 works in our library](https://md.research.elamur.ai/free/en/authors/rahul-k-arora/#papers) · [643 citations in our library](https://md.research.elamur.ai/free/en/authors/rahul-k-arora/cited-by/) · [⭐ Paper Star rank #2544](https://md.research.elamur.ai/free/en/paper-stars/page/26/#star-2544) The profile is the library's match of the name; two people with one name are not always told apart. Elsewhere: [Google Scholar](https://scholar.google.com/citations?view_op=search_authors&mauthors=Rahul+K.+Arora) · [Semantic Scholar](https://www.semanticscholar.org/search?q=Rahul+K.+Arora) · [dblp](https://dblp.org/search?q=Rahul+K.+Arora) · [ACL Anthology](https://aclanthology.org/search/?q=Rahul+K.+Arora) ## Topics [Language models & pretraining 2](https://md.research.elamur.ai/free/en/topics/language-models/) · [Reasoning 1](https://md.research.elamur.ai/free/en/topics/reasoning/) ## Works by year ### 2020s [**2025** 2 papers](https://md.research.elamur.ai/free/en/authors/rahul-k-arora/year/2025/) ## Papers The most cited first (OpenAlex and our library). 1. **[gpt-oss-120b & gpt-oss-20b Model Card](https://md.research.elamur.ai/free/en/papers/agarwal-2025-gpt-oss-120b-gpt-oss/)** *(2025 · 555 citations)* This model card from OpenAI describes gpt-oss-120b and gpt-oss-20b. They are text-only reasoning models with open weights under the Apache 2.0 license. They are built for agentic workflows: web search, Python execution, function calling and Structured Outputs. Users can set how much effort the model spends on… 2. **[HealthBench: Evaluating Large Language Models Towards Improved Human Health](https://md.research.elamur.ai/free/en/papers/arora-2025-healthbench-evaluating-large-language-models/)** *(2025 · 115 citations)* Most medical benchmarks for LLMs use multiple-choice or short-answer questions. These capture little of how people actually seek health help, are rarely checked against expert judgment, and are often close to saturated. HealthBench, from OpenAI, addresses this with 5,000 realistic single- and multi-turn health… ## Co-authors here 8 co-authors with a page, the most papers together first. - [Alex Beutel](https://md.research.elamur.ai/free/en/authors/alex-beutel/) - [Bowen Baker](https://md.research.elamur.ai/free/en/authors/bowen-baker/) - [Jason Wei](https://md.research.elamur.ai/free/en/authors/jason-wei/) - [Johannes Heidecke](https://md.research.elamur.ai/free/en/authors/johannes-heidecke/) - [Karan Singhal](https://md.research.elamur.ai/free/en/authors/karan-singhal/) - [Lama Ahmad](https://md.research.elamur.ai/free/en/authors/lama-ahmad/) - [Sam Altman](https://md.research.elamur.ai/free/en/authors/sam-altman/) - [Yu Bai](https://md.research.elamur.ai/free/en/authors/yu-bai/) ## Cited by Papers of our library cite their works 643 times (of the 140 884 papers whose bibliographies we read); [33 citing papers with a page here →](https://md.research.elamur.ai/free/en/authors/rahul-k-arora/cited-by/).