Multimodal ModelsEvaluation & RobustnessAI Safety
I build and evaluate language and multimodal models, with a focus on reasoning, robustness, and safety.
I'm looking for a full-time Research Scientist role in industry, working on AI safety, evaluations, and multimodal models. Based in the UK · open to UK-based or remote roles · get in touch
About
I'm a Postdoctoral Researcher at the University of Edinburgh, working with Pasquale Minervini on long-context language and vision-language models, with an emphasis on evaluation, robustness, reasoning, and AI safety. I completed my PhD in NLP at Edinburgh (EdinburghNLP, CDT NLP), advised by Frank Keller and Hao Tang.
Most recently I was a Research Fellow at Anthropic, studying model behaviour and dual-use risks in frontier LLMs: how seemingly benign generations can be repurposed for harm, and what that means for evaluation and deployment safeguards. I also co-mentor a SPAR project on multimodal safety and mechanistic interpretability in encoder-free vision-language models.
Previously, I was an Applied Science intern at AWS AI, working on hallucination mitigation in LLMs. Outside research, I enjoy photography.
Research Work in progress →
Multimodal models
How vision-language models understand images, video and documents, and where they fall short, from reading clocks to summarizing scientific posters.
Evaluation & robustness
Benchmarks that expose where vision-language models break: corruptions, long context, and inverse scaling.
AI safety & misuse
Dual-use risks in frontier LLMs, and safety evaluation of LLM agents.
Long-context & reasoning
Long-document understanding with memory-efficient end-to-end training, and whether chain-of-thought reflects what actually drives a model's answer.
In the media
Lost in Time, our study of clock and calendar understanding in multimodal LLMs, was cited in Stanford HAI's 2026 AI Index Report and covered internationally.
News
- 2026Do Composed Image Retrieval Benchmarks Require Multimodal Composition? accepted at NeurIPS 2026 Datasets & Benchmarks.
- 2026VLM-RobustBench accepted at ICML 2026.
- 2026Lost in Time cited in Stanford HAI's 2026 AI Index Report.
- May 2026Completed the Anthropic Fellows program.
- Jan 2026Completed my PhD at the University of Edinburgh.
- Nov 2025Joined Anthropic as a Research Fellow in London.
- Sep 2025MMLongBench accepted at NeurIPS 2025 Datasets & Benchmarks Spotlight, plus one NeurIPS workshop paper.
- 2025PosterSum accepted at AACL 2025.
Selected publications Scholar ↗
-
On the Limits of Scale: Inverse Scaling in Vision-Language Models
Rohit Saxena, Alessandro Suglia, Pasquale Minervini
Under review -
Aryo Pradipta Gema, Neel Rajani, Rohit Saxena, Wai-Chung Kwan, Pasquale Minervini
PreprintarXiv ↗ -
Do Composed Image Retrieval Benchmarks Require Multimodal Composition?
Matteo Attimonelli, Alessandro De Bellis, Aryo Pradipta Gema, Rohit Saxena, et al.
NeurIPS 2026 D&BarXiv ↗ -
VLM-RobustBench: A Comprehensive Benchmark for Robustness of Vision-Language Models
Rohit Saxena, Alessandro Suglia, Pasquale Minervini
ICML 2026arXiv ↗ -
Lost in Time: Clock and Calendar Understanding Challenges in Multimodal LLMs
Rohit Saxena, Aryo Pradipta Gema, Pasquale Minervini
-
MMLongBench: Benchmarking Long-Context Vision-Language Models Effectively and Thoroughly
Zhaowei Wang, Wenhao Yu, Xiyu Ren, Jipeng Zhang, Yu Zhao, Rohit Saxena, et al.
-
End-to-End Long Document Summarization using Gradient Caching
Rohit Saxena, Hao Tang, Frank Keller
TACL 2025Paper ↗ -
PosterSum: A Multimodal Benchmark for Scientific Poster Summarization
Rohit Saxena, Pasquale Minervini, Frank Keller
AACL 2025arXiv ↗
All publications
On the Limits of Scale: Inverse Scaling in Vision-Language Models
Rohit Saxena, Alessandro Suglia, Pasquale Minervini
Under reviewAryo Pradipta Gema, Neel Rajani, Rohit Saxena, Wai-Chung Kwan, Pasquale Minervini
PreprintDo Composed Image Retrieval Benchmarks Require Multimodal Composition?
Matteo Attimonelli, Alessandro De Bellis, Aryo Pradipta Gema, Rohit Saxena, et al.
NeurIPS 2026 D&BVLM-RobustBench: A Comprehensive Benchmark for Robustness of Vision-Language Models
Rohit Saxena, Alessandro Suglia, Pasquale Minervini
ICML 2026MMLongBench: Benchmarking Long-Context Vision-Language Models Effectively and Thoroughly
Zhaowei Wang, Wenhao Yu, Xiyu Ren, Jipeng Zhang, Yu Zhao, Rohit Saxena, et al.
NeurIPS 2025 D&B · SpotlightEnd-to-End Long Document Summarization using Gradient Caching
Rohit Saxena, Hao Tang, Frank Keller
TACL 2025PosterSum: A Multimodal Benchmark for Scientific Poster Summarization
Rohit Saxena, Pasquale Minervini, Frank Keller
AACL 2025What Is That Talk About? A Video-to-Text Summarization Dataset for Scientific Presentations
Dongqi Liu, Chenxi Whitehouse, Xi Yu, Louis Mahon, Rohit Saxena, et al.
ACL 2025Lost in Time: Clock and Calendar Understanding Challenges in Multimodal LLMs
Rohit Saxena, Aryo Pradipta Gema, Pasquale Minervini
ICLR 2025 Workshop on Reasoning and Planning for LLMsAryo Pradipta Gema, Joshua Ong, Giwon Hong, Alessio Devoto, Alberto Mancino, Rohit Saxena, et al.
NAACL 2025Select and Summarize: Scene Saliency for Movie Script Summarization
Rohit Saxena, Frank Keller
Findings of NAACL 2024MovieSum: An Abstractive Summarization Dataset for Movie Screenplays
Rohit Saxena, Frank Keller
Findings of ACL 2024The Hallucinations Leaderboard: An Open Effort to Measure Hallucinations in Large Language Models
Giwon Hong*, Aryo Pradipta Gema*, Rohit Saxena*, et al. (*equal contribution)
PreprintRohit Saxena, Savita Bhat, Niranjan Pedanekar
SocialNLP @ ACL 2018Rohit Saxena, Savita Bhat, Niranjan Pedanekar
ICDM Workshops 2017I Know What You Coded Last Summer: Mining Candidate Expertise from GitHub Repositories
Rohit Saxena, Niranjan Pedanekar
CSCW 2017 Companion
Patents
- US10599864B2: Sensitive audio zone rearrangement for customer verification.
- US10296523B2: System and method for estimating temporal importance of data.
- US10198322B2: Method and system for efficient selective backup strategy in an enterprise.
- US20150381703A1: Automating a process for web-based software.
- US20160269417A1: Dynamic data masking for mainframe application.
- 201621003887 (IN): Systems and methods for estimating skill-sets of users in a distributed environment.
Contact
© 2026 Rohit Saxena