Mohit Iyyer
Associate Professor
Studies large language models for long-context understanding, long-form generation evaluation, factuality and literary translation.
Mohit Iyyer's lab at UMD works on large language models and interpretability, evaluation & safety of LMs. 47 papers since 2023, 25 at top venues, mostly EMNLP and ACL. 13 alumni, most often to Google DeepMind and Amazon.
At a glance
- papers, 2023–2026
- 12/ year
- at T1 venues
- 53%
- PhD students
- 0
- MS students
- 0
- alumni listed
- 13
Papers by yearfrom the lab homepage, 2023–2026
- 9 papers2026 · T1 7 · collecting
- 13 papers2025 · T1 7
- 14 papers2024 · T1 6
- 11 papers2023 · T1 5
T1 venuesOtherVenues: EMNLP, ACL, Findings of EMNLP, Findings of ACL, EACL
Research focuspaper titles that use each topic's words
- Large language models20 papers
- long-context5 papers
- Interpretability, evaluation & safety of LMs4 papers
- factuality2 papers
- studies large0 papers
Paper list
From the lab homepage.
20269 papers
- ACL (T1)AI use in American newspapers is widespread, uneven, and rarely disclosed (opens in a new tab)
- Findings of ACLAnalystBench: Benchmarking professional long-form report generation with web-mined multimodal tasks (opens in a new tab)
- EMNLP (T1)Argument Collapse: LLMs Flatten Long-Form Public Debate (opens in a new tab)
- ICML (T1)Bio-Inspired Self-Supervised Learning for Wrist-worn IMU Signals (opens in a new tab)
- ICLR (T1)EditLens: Quantifying the Extent of AI Editing in Text (opens in a new tab)
- ACL (T1)Frankentext: Stitching random text fragments into long-form narratives (opens in a new tab)
- ICML (T1)How2Everything: Mining the Web for How-To Procedures to Evaluate and Improve LLMs (opens in a new tab)
- EMNLP (T1)POLARIS: Guiding Small Models to Write Long Stories (opens in a new tab)
- COLMStoryScope: Investigating idiosyncrasies in AI fiction (opens in a new tab)
202513 papers
- COLMBEARCUBS: A benchmark for computer-using web agents (opens in a new tab)
- NeurIPS (T1)BLEUBERI: BLEU is a surprisingly effective reward for instruction following (opens in a new tab)
- ACL (T1)CaLMQA: Exploring culturally specific long-form question answering across 23 languages (opens in a new tab)
- COLMCLIPPER: Compression enables long-context synthetic data generation (opens in a new tab)
- TACL (T1)Contextualized Evaluations: Judging Language Model Responses to Underspecified Queries (opens in a new tab)
- EMNLP (T1)Does quantization affect models' performance on long-context tasks? (opens in a new tab)
- ACL (T1)Literary Evidence Retrieval via Long-Context Language Models (opens in a new tab)
- Findings of ACLLocalizing and Mitigating Errors in Long-form Question Answering (opens in a new tab)
- COLMOne ruler to measure them all: Benchmarking multilingual long-context language models (opens in a new tab)
- EMNLP (T1)OWL: Probing Cross-Lingual Recall of Memorized Texts via World Literature (opens in a new tab)
- ACL (T1)People who frequently use ChatGPT for writing tasks are accurate and robust detectors of AI-generated text (opens in a new tab)
- Findings of EMNLPVeriFastScore: Speeding up long-form factuality evaluation (opens in a new tab)
202414 papers
- ICLR (T1)BooookScore: A systematic exploration of book-length summarization in the era of LLMs (opens in a new tab)
- COLMFABLES: Evaluating faithfulness and content selection in book-length summarization (opens in a new tab)
- Findings of ACLFreshLLMs: Refreshing Large Language Models with Search Engine Augmentation (opens in a new tab)
- Findings of NAACLGEE! Grammar Error Explanation with Large Language Models (opens in a new tab)
- Findings of EACLHow Does In-Context Learning Help Prompt Tuning? (opens in a new tab)
- COLMIteratively Prompting Multimodal LLMs to Reproduce Natural and AI-Generated Images (opens in a new tab)
- ACL (T1)Multistage Collaborative Knowledge Distillation from a Large Language Model for Semi-Supervised Sequence Generation (opens in a new tab)
- EMNLP (T1)One Thousand and One Pairs: A "novel" challenge for long-context language models (opens in a new tab)
- EACLPEARL: Prompting Large Language Models to Plan and Execute Actions Over Long Documents (opens in a new tab)
- EMNLP (T1)PostMark: A Robust Blackbox Watermark for Large Language Models (opens in a new tab)
- Findings of EMNLPSuri: Multi-constraint Instruction Following for Long-form Text Generation (opens in a new tab)
- NAACL (T1)TopicGPT: A Prompt-based Topic Modeling Framework (opens in a new tab)
202311 papers
- ACL (T1)A Critical Evaluation of Evaluations for Long-form Question Answering (opens in a new tab)
- Findings of EMNLPDisco Elysium: Exploring Player Perceptions of LLM-Generated Dialogue within a Commercial Video Game (opens in a new tab)
- Findings of EACLezCoref: Towards Unifying Annotation Guidelines for Coreference Resolution (opens in a new tab)
- EMNLP (T1)FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation (opens in a new tab)
- EMNLP (T1)kNN-LM Does Not Improve Open-ended Text Generation (opens in a new tab)
- WMTLarge language models effectively leverage document-level context for literary translation, but critical errors persist (opens in a new tab)
- EACLLongEval: Guidelines for Human Evaluation of Faithfulness in Long-form Summarization (opens in a new tab)
- Findings of EMNLPPaRaDe: Passage Ranking using Demonstrations with LLMs (opens in a new tab)
- NeurIPS (T1)Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defense (opens in a new tab)
- CCS (T1)Stealing the Decoding Algorithms of Language Models (opens in a new tab)
- Workshop on Creative AI Across Modalities @ AAAITowards Grounded Dialogue Generation in Video Game Environments (opens in a new tab)
Where alumni went13 listed, 77% to industry
Industry 10Academia 3
Most often to
- Google DeepMind
- Amazon
- at Walmart Labs
Also to
- Cornell University
Apply to UMDFall 2027
Not announced yet
UMD Computer Science PhD
- Round
- Single round
- Opens
- —
- GRE
- See program page
- English
- See program page
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- USStanford
Daniel E. Ho
- Large language models
- USUCSD
Ndapa Nakashole
- Knowledge, QA & retrieval
- USCMU
Eric Nyberg
- Reasoning & agents
- USGeorgia Tech
Anna Ivanova
- Large language models
- GBCambridge
Ann Copestake
- Dialogue, generation & summarization
- USCMU
Graham Neubig
- Large language models
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.