Elamur Research⌕
☰
home / authors / Ashish Panwar

Ashish Panwar

Ashish Panwar has 3 papers in our library, in 2023–2024, most of them on Language models & pretraining; often with Amey Agrawal, Bhargav S. Gulavani, Jayashree Mohan.

3 papers in our library · 2023–2024

OpenAlex · ORCID · h-index 11 · 21 works · 492 citations OpenAlex, 2026-10-14

📄 3 works in our library · 🔗 Cited 564 times in our library · ⭐ Paper Stars ranking: #3819

The profile is the library's match of the name; two people with one name are not always told apart.

Elsewhere: Google Scholar · Semantic Scholar · dblp · ACL Anthology

 Affiliations

🏢Microsoft ResearchBengaluru · India2023–2026 · 8 paperscurrent🎓Indian Institute of Science BangaloreBengaluru · India2015–2025 · 9 papers🏢MicrosoftRedmond · United States2022–2024 · 1 paperEmployment since 2022 · Senior Researcher

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 1 more institution on single bylines are not listed.

 Fields, topics and concepts

Of its works in our library, with the number of works on each.

📚 Shelves 7The field of science this work belongs to (OpenAlex fields).Computer Science · 3 worksArts and Humanities · 2 worksHardware and Architecture · 3 worksArtificial Intelligence · 2 worksLanguage and Linguistics · 2 worksComputer Networks and Communications · 1 workInformation Systems · 1 work

🧭 Topics 1Research subjects and languages the work is about.Language models & pretraining · 3 works

 Works by year

2020s

20242 papers20231 paper

 Papers

The most cited first (OpenAlex and our library).

  1. Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve (2024 · 369 citations)
    Serving LLMs online requires both high throughput, to keep costs down, and low latency between generated tokens.
  2. SARATHI: Efficient LLM Inference by Piggybacking Decodes with Chunked Prefills (2023 · 144 citations)
    Serving large language models is inefficient because generation has two very different stages. Processing the prompt (prefill) keeps a GPU fully busy even with a single request. Producing output one token at a time (decode) leaves most compute idle unless batches are large. Under pipeline parallelism, micro-batches…
  3. Vidur: A Large-Scale Simulation Framework For LLM Inference (2024 · 90 citations)
    Deploying a large language model involves many choices: the parallelism layout (TP, PP, number of replicas), the scheduler (Orca, vLLM, Sarathi-Serve), batch size, scheduler-specific knobs and the GPU SKU.

 Co-authors here

7 co-authors with a page, the most papers together first.

Venues

Where their works here were published, the most works first, with the editions of those that have a page.

 Cited by

Papers of our library cite their works 564 times (of the 149,439 papers whose bibliographies we read); 22 citing papers with a page here →.

Updated

This page in Markdown → https://md.research.elamur.ai/free/en/authors/ashish-panwar/