RM
Rada Mihalcea
cs.CLcs.AIcs.LGcs.CYcs.CVcs.HCcs.SDeess.ASCOVID-19cs.SI
On Valency
published · living versionsW_gb4p66m9·v1 · currentpublished
The Culture Within: Towards Mechanistic Understanding of Cultural Reasoning in LLMs
with Samee Arif, Furkan Danisman, Angana Borah, Honglak Lee +4
1 version
W_t6vmu6zg·v1published
Preprints & journals
162 papers in the corpus · 2011–2026SWE-VERIFIER: Selective Verification for Cost-Efficient Coding Agents2610.30284v1 · Kaijian Zou, Vaibhav Chennareddy, Yunxiang Zhang et al.2026 · 0 citationsarXiv
The Culture Within: Towards Mechanistic Understanding of Cultural Reasoning in LLMs2610.27202v1 · Samee Arif, Furkan Danisman, Angana Borah et al.2026 · 0 citationsarXivon Valency
AtlasNLP: A Country-Aware Atlas of Dataset Representation in NLP2608.30107v2 · Joan Nwatu, Tsedeniya Solomon Amare, Longju Bai et al.2026 · 0 citationsarXiv
A reporting checklist for large language models in behavioural science.42265331 · Feuerriegel, Stefan, Barrie, Christopher, Crockett, M J et al.2026 · 16 citationsNature human behaviour. 2026;10(7):1182-1186
Wait, am I Being Fair? Characterizing Deductive Stereotyping and Mitigating It with Fair-GCG2606.30989v1 · Naihao Deng, Yilun Zhu, Joan Nwatu et al.2026 · 0 citationsarXiv
The Language-Energy Divide: Measuring Energy Costs of Multilingual LLM Inference2606.21869v2 · Naihao Deng, Alissa Shen, Yiming Feng et al.2026 · 0 citationsarXiv
The Wrong Kind of Right: Quantifying and Localizing Misfired Alignment in LLMs2606.18656v1 · Naihao Deng, Yiming Feng, Chimaobi Okite et al.2026 · 0 citationsarXiv
It Takes One to Bias Them All: Breaking Bad with One-Shot GRPO2606.10931v3 · Naihao Deng, Yilun Zhu, Naichen Shi et al.2026 · 0 citationsarXiv
Whose Norms? Disentangling Cultural and Personal Alignment in Large Language Models2606.07877v1 · Angana Borah, Isabelle Augenstein, Rada Mihalcea2026 · 0 citationsarXiv
The Age of Curiosity Meets the Age of AI: Benchmarking Child Safety in Large Language Models2605.25510v3 · Samee Arif, Angana Borah, Rada Mihalcea2026 · 0 citationsarXiv
How Do AI Agents Spend Your Money? Analyzing and Predicting Token Consumption in Agentic Coding Tasks2604.22750v3 · Longju Bai, Zhemin Huang, Xingyao Wang et al.2026 · 3 citationsarXiv
SafetyALFRED: Evaluating Safety-Conscious Planning of Multimodal Large Language Models2604.19638v1 · Josue Torres-Fonseca, Naihao Deng, Yinpei Dai et al.2026 · 0 citationsarXiv
DOTRAG: Retrieval-Time Reasoning Along Paths2605.18760v1 · Larnell Moore, Naihao Deng, Rada Mihalcea et al.2026 · 0 citationsarXiv
One Word at a Time: Incremental Completion Decomposition Breaks LLM Safety2604.25921v2 · Samee Arif, Naihao Deng, Zhijing Jin et al.2026 · 0 citationsarXiv
When Do Language Models Endorse Limitations on Human Rights Principles?2603.04217v1 · Keenan Samway, Nicole Miu Takagi, Rada Mihalcea et al.2026 · 0 citationsarXiv
Culture Affordance Atlas: Reconciling Object Diversity Through Functional Mapping2512.03173v1 · Joan Nwatu, Longju Bai, Oana Ignat et al.2025 · 0 citationsAAAI 2026 Social Impact Track
Are LLMs Good Safety Agents or a Propaganda Engine?2511.23174v1 · Neemesh Yadav, Francesco Ortu, Jiarui Liu et al.2025 · 0 citationsarXiv
Evaluating the reliability of large language models for clinical data extraction in bladder cancer prognosis.41272207 · Sun, Di, Hadjiiski, Lubomir, Bruno, Grace et al.2025 · 9 citationsScientific reports. 2025;15(1):43773
SocialHarmBench: Revealing LLM Vulnerabilities to Socially Harmful Requests2510.04891v2 · Punya Syon Pandey, Hai Son Le, Devansh Bhardwaj et al.2025 · 0 citationsarXiv
Benchmarking and Improving LLM Robustness for Personalized Generation2509.19358v1 · Chimaobi Okite, Naihao Deng, Kiran Bodipati et al.2025 · 0 citationsFindings of the Association for Computational Linguistics: EMNLP 2025, 2025
ISCA: A Framework for Interview-Style Conversational Agents2508.14344v1 · Charles Welch, Allison Lahnala, Vasudha Varadarajan et al.2025 · 0 citationsarXiv
MOMENTS: A Comprehensive Multimodal Benchmark for Theory of Mind2507.04415v2 · Emilio Villa-Cueva, S M Masrur Ahmed, Rendi Chevi et al.2025 · 1 citationarXiv
Patient-Centered RAG for Oncology Visit Aid Following the Ottawa Decision Guide2507.04026v1 · Siyang Liu, Lawrence Chin-I An, Rada Mihalcea2025 · 0 citationsarXiv
Now More Than Ever, Foundational AI Research and Infrastructure Depends on the Federal Government2506.14679v1 · Michela Taufer, Rada Mihalcea, Matthew Turk et al.2025 · 0 citationsarXiv
Which Humans? Inclusivity and Representation in Human-Centered AI2506.14680v1 · Rada Mihalcea, Nazanin Andalibi, David Jensen et al.2025 · 0 citationsarXiv
Democratic or Authoritarian? Probing a New Dimension of Political Biases in Large Language Models2506.12758v2 · David Guzman Piedrahita, Irene Strauss, Bernhard Scholkopf et al.2025 · 1 citationarXiv
CliniDial: A Naturally Occurring Multimodal Dialogue Dataset for Team Reflection in Action During Clinical Operation2506.12936v1 · Naihao Deng, Kapotaksha Das, Rada Mihalcea et al.2025 · 0 citationsarXiv
Free Lunch for User Experience: Crowdsourcing Agents for Scalable User Studies2505.22981v2 · Siyang Liu, Sahand Sabour, Xiaoyang Wang et al.2025 · 0 citationsarXiv
Are Language Models Consequentialist or Deontological Moral Reasoners?2505.21479v2 · Keenan Samway, Max Kleiman-Weiner, David Guzman Piedrahita et al.2025 · 2 citationsarXiv
When Ethics and Payoffs Diverge: LLM Agents in Morally Charged Social Dilemmas2505.19212v2 · Steffen Backmann, David Guzman Piedrahita, Terry Jingchen Zhang et al.2025 · 0 citationsarXiv
Evaluation Framework for AI Systems in "the Wild"2504.16778v2 · Sarah Jabbour, Trenton Chang, Anindya Das Antar et al.2025 · 1 citationarXiv
Revealing Hidden Mechanisms of Cross-Country Content Moderation with Natural Language Processing2503.05280v2 · Neemesh Yadav, Jiarui Liu, Francesco Ortu et al.2025 · 2 citationsarXiv
Mind the (Belief) Gap: Group Identity in the World of LLMs2503.02016v2 · Angana Borah, Marwa Houalla, Rada Mihalcea2025 · 1 citationarXiv
Persuasion at Play: Understanding Misinformation Dynamics in Demographic-Aware Human-LLM Interactions2503.02038v2 · Angana Borah, Rada Mihalcea, Ver'onica P'erez-Rosas2025 · 1 citationarXiv
Eeyore: Realistic Depression Simulation via Supervised and Preference Optimization2503.00018v1 · Siyang Liu, Bianca Brie, Wenda Li et al.2025 · 0 citationsarXiv
Analysis of social media language reveals the psychological interaction of three successive upheavals.39962124 · Mesquiti, Steven, Seraj, Sarah, Weyland, Andreas H et al.2025 · 11 citationsScientific reports. 2025;15(1):5740
Examining Spanish Counseling with MIDAS: a Motivational Interviewing Dataset in Spanish2502.08458v1 · Aylin Gunal, Bowen Yi, John Piette et al.2025 · 1 citationarXiv
Human Decision-making is Susceptible to AI-driven Manipulation2502.07663v3 · Sahand Sabour, June M. Liu, Siyang Liu et al.2025 · 6 citationsarXiv
Are Human Interactions Replicable by Generative Agents? A Case Study on Pronoun Usage in Hierarchical Interactions2501.15283v1 · Naihao Deng, Rada Mihalcea2025 · 0 citationsarXiv
Rethinking Table Instruction Tuning2501.14693v4 · Naihao Deng, Rada Mihalcea2025 · 1 citationarXiv
Chumor 2.0: Towards Benchmarking Chinese Humor Understanding2412.17729v1 · Ruiqi He, Yushu He, Longju Bai et al.2024 · 0 citationsarXiv
The Power of Many: Multi-Agent Multimodal Models for Cultural Image Captioning2411.11758v1 · Longju Bai, Angana Borah, Oana Ignat et al.2024 · 2 citationsarXiv
Why AI Is WEIRD and Should Not Be This Way: Towards AI For Everyone, With Everyone, By Everyone2410.16315v1 · Rada Mihalcea, Oana Ignat, Longju Bai et al.2024 · 4 citationsarXiv
Towards Implicit Bias Detection and Mitigation in Multi-Agent LLM Interactions2410.02584v1 · Angana Borah, Rada Mihalcea2024 · 15 citationsarXiv
How developments in natural language processing help us in understanding human behaviour.39438680 · Mihalcea, Rada, Biester, Laura, Boyd, Ryan L et al.2024 · 30 citationsNature human behaviour. 2024;8(10):1877-1889
Language Model Alignment in Multilingual Trolley Problems2407.02273v6 · Zhijing Jin, Max Kleiman-Weiner, Giorgio Piatti et al.2024 · 1 citationarXiv
Uplifting Lower-Income Data: Strategies for Socioeconomic Perspective Shifts in Large Multi-modal Models2407.02623v3 · Joan Nwatu, Oana Ignat, Rada Mihalcea2024 · 0 citationsarXiv
Outcome Prediction Using Multi-Modal Information: Integrating Large Language Model-Extracted Clinical Information and Image Analysis.39001463 · Sun, Di, Hadjiiski, Lubomir, Gormley, John et al.2026 · 30 citationsCancers. 2024;16(13)
Towards Region-aware Bias Evaluation Metrics2406.16152v2 · Angana Borah, Aparna Garimella, Rada Mihalcea2024 · 0 citationsarXiv
CVQA: Culturally-diverse Multilingual Visual Question Answering Benchmark2406.05967v2 · David Romero, Chenyang Lyu, Haryo Akbarianto Wibowo et al.2024 · 6 citationsarXiv
Career total: 566 works. 162 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.