Rada Mihalcea
Janice M. Jenkins Collegiate Professor of Computer Science and Engineering; Professor, EECS – Computer Science and Engineering; Director, Artificial Intelligence Lab
Studies natural language processing and computational social science, including multimodal understanding of human behavior.
Rada Mihalcea's lab at Michigan works on language & society, multimodal & grounded language and vision-language & multimodal. 108 papers since 2023, 26 at top venues, mostly EMNLP and Findings of EMNLP. 12 PhD students and 4 master's students. 102 alumni, most often to Google and IBM. Applications close .
At a glance
- papers, 2023–2026
- 28/ year
- at T1 venues
- 24%
- PhD students
- 12
- MS students
- 4
- alumni listed
- 102
Papers by yearfrom the lab homepage, 2023–2026
- 23 papers2026 · T1 4 · collecting
- 41 papers2025 · T1 8
- 19 papers2024 · T1 10
- 25 papers2023 · T1 4
T1 venuesOtherVenues: EMNLP, Findings of EMNLP, Findings of ACL, COLING, EACL
Research focuspaper titles that use each topic's words
- Multimodal & grounded language32 papers
- Vision-language & multimodal32 papers
- Language & society31 papers
- human behavior0 papers
- social science0 papers
Paper list
From the lab homepage.
202623 papers
- —A reporting checklist for large language models in behavioural science. (opens in a new tab)
- —AI poses risks to democratic and social systems (opens in a new tab)
- EMNLP (T1)AtlasNLP: A Country-Aware Atlas of Dataset Representation in NLP (opens in a new tab)
- —Belief-Sim: Towards Belief-Driven Simulation of Demographic Misinformation Susceptibility (opens in a new tab)
- ACL (T1)Copyright Detective: A Forensic System to Evidence LLMs Flickering Copyright Leakage Risks (opens in a new tab)
- —DOTRAG: Retrieval-Time Reasoning Along Paths (opens in a new tab)
- ACM Transactions on Computing for HealthcareEvaluating Language Models for Assessing Counselor Reflections (opens in a new tab)
- —How do coding agents spend your money? Analyzing and predicting token consumptions in agentic coding tasks (opens in a new tab)
- —It Takes One to Bias Them All: Breaking Bad with One-Shot GRPO (opens in a new tab)
- —LUCid: Redefining Relevance For Lifelong Personalization (opens in a new tab)
- IEEE MultiMediaMultimodal Non-Contact Based Modeling of Enervation (opens in a new tab)
- EACLNLP for social good: A survey and outlook of challenges, opportunities and responsible deployment (opens in a new tab)
202541 papers
- ACL (T1)Acoustic Individual Identification of White-Faced Capuchin Monkeys Using Joint Multi-Species Embeddings (opens in a new tab)
- —Are Human Interactions Replicable by Generative Agents? A Case Study on Pronoun Usage in Hierarchical Interactions (opens in a new tab)
- —Are Language Models Consequentialist or Deontological Moral Reasoners? (opens in a new tab)
- —Are LLMs Good Safety Agents or a Propaganda Engine? (opens in a new tab)
- Findings of EMNLPBenchmarking and Improving LLM Robustness for Personalized Generation (opens in a new tab)
- Findings of ACLChumor 2.0: Towards better benchmarking chinese humor understanding from (ruo zhi ba) (opens in a new tab)
- —CliniDial: A Naturally Occurring Multimodal Dialogue Dataset for Team Reflection in Action During Clinical Operation (opens in a new tab)
- AAAI (T1)Culture Affordance Atlas: Reconciling Object Diversity Through Functional Mapping (opens in a new tab)
- EACLDemocratic or Authoritarian? Probing a New Dimension of Political Biases in Large Language Models (opens in a new tab)
- Findings of ACLEeyore: Realistic Depression Simulation via Expert-in-the-Loop Supervised and Preference Optimization (opens in a new tab)
- Findings of ACLEvaluating LLMs’ Mathematical and Coding Competency through Ontology-guided Interventions (opens in a new tab)
- —Evaluation Framework for AI Systems in" the Wild" (opens in a new tab)
202419 papers
- COLING (T1)A Language Model Trained on Uruguayan Spanish News Text (opens in a new tab)
- COLING (T1)Annotations on a Budget: Leveraging Geo-Data Similarity to Balance Model Performance and Annotation Cost (opens in a new tab)
- WACVAugment the Pairs: Semantics-Preserving Image-Caption Pair Augmentation for Grounding-Based Vision and Language Models (opens in a new tab)
- ICLR (T1)Can Large Language Models Infer Causation from Correlation? (opens in a new tab)
- —CloVe: Encoding Compositional Language in Contrastive Vision-Language Models (opens in a new tab)
- NeurIPS (T1)CVQA: Culturally-diverse Multilingual Visual Question Answering Benchmark (opens in a new tab)
- COLING (T1)Dynamic Reward Adjustment in Multi-Reward Reinforcement Learning for Counselor Reflection Generation (opens in a new tab)
- COLING (T1)Has It All Been Solved? Open NLP Research Questions Not Solved by Large Language Models (opens in a new tab)
- Nature Human BehaviourHow developments in natural language processing help us in understanding human behaviour (opens in a new tab)
- TextGraphs-17Human Action Co-occurrence in Lifestyle Vlogs using Graph Link Prediction (opens in a new tab)
- —Quriosity: Analyzing Human Questioning Behavior And Causal Inquiry through Curiosity-Driven Queries (opens in a new tab)
- —SQL-CRAFT: Text-to-SQL through Interactive Refinement and Enhanced Reasoning (opens in a new tab)
202325 papers
- Information FusionA Review of Deep Learning Techniques for Speech Processing (opens in a new tab)
- Findings of EMNLPBeyond Good Intentions: Reporting the Research Landscape of NLP for Social Good (opens in a new tab)
- EMNLP (T1)Bridging the Digital Divide: Performance Variation across Socio-Economic Factors in Vision-Language Models (opens in a new tab)
- EMNLP (T1)Cross-Cultural Analysis of Human Values, Morals, and Biases in Folk Tales (opens in a new tab)
- International Conference on the Science of Science and InnovationEditing a Woman’s Voice (opens in a new tab)
- —Emergent Linear Representations in World Models of Self-Supervised Sequence Models (opens in a new tab)
- EACLEmpathy Identification Systems are not Accurately Accounting for Context (opens in a new tab)
- IEEE International Conference on Acoustics, Speech and Signal ProcessingEvaluating Parameter-Efficient Transfer Learning Approaches on SURE Benchmark for Speech Understanding (opens in a new tab)
- Findings of EMNLPHi-ToM: A Benchmark for Evaluating Higher-Order Theory of Mind Reasoning in Large Language Models (opens in a new tab)
- Proceedings of the International AAAI Conference on Web and Social MediaImproving Mental Health Classifier Generalization with Pre-Diagnosis Data (opens in a new tab)
- Findings of EMNLPLanguage Guided Visual Question Answering: Elevate Your Multimodal Language Model Using Knowledge-Enriched Prompts (opens in a new tab)
- ASEELexical Measurement of Teaching Qualities (opens in a new tab)
Where alumni went102 listed, 50% to industry
Industry 10Academia 10
Most often to
- IBM
- Netflix
Also to
- University of Michigan
- Caltech
- Middlebury College
Apply to MichiganFall 2027
Michigan CSE 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
- GBEdinburgh
Emily Allaway
- Large language models
- USNYU
Eunsol Choi
- Knowledge, QA & retrieval
- USNYU
Tal Linzen
- Multilingual, translation & linguistics
- CNPeking
Yansong Feng
- Knowledge, QA & retrieval
- SGNTU
Erik Cambria
- Reasoning & agents
- CNPeking
Houfeng Wang
- Knowledge, QA & retrieval
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.
Michigancloses