Xuhui Zhou
GHC 5705
4902 Forbes Ave
Pittsburgh, PA 15213
I am a PhD student at the Language Technologies Institute at CMU fortunately advised by Maarten Sap . I am interested in socially intelligent AI. More specifically, I am interested in the following Qs:
- How do we define and build socially intelligent AI systems? e.g., Sotopia
- How do we create better (socially) grounded AI systems? e.g., WebArena
- How do we safegaurd AI systems from harmful behaviors? e.g., HAICOSYSTEM
news
Nov 4, 2024 | Hi there! Iβm excited to find a summer 2025 internship and would love to hear from you if you think I could be a great match for your team. Feel free to drop me a message on X (nlpxuhui) or send an email to xuhuiz@cs.cmu.edu. Looking forward to connecting! |
---|---|
Oct 22, 2024 | π π HAICOSYSTEM is now out! https://arxiv.org/abs/2409.16427 π Website π Tweet |
Oct 19, 2024 | π£οΈ π’ Invited talk about Towards Socially Aware and Safe AI Agents at MIT and USC. |
Aug 1, 2024 | π Consent in Crisis: The Rapid Decline of the AI Data Commons is finally out! π Website ποΈ New Yor Times π Tweet |
Jun 1, 2024 | π β΅ Started my internship at AI2 Mosaic Team. Also PolygloToxicityPrompts: Multilingual Evaluation of Neural Toxic Degeneration in Large Language Models got accepted at COLM 2024 (π Tweet) π |
Apr 19, 2024 | π£οΈ π’ Gave a talk about Towards Socially Aware and Interactional NLP Systems at CMU Foundation and Language Model Seminar and MilaNLP. |
Mar 2, 2024 | I am attending ICLR from May 7th to May 10th. Please reach out and letβs chat about socially-aware AI and alignment! |
Jan 16, 2024 | Really excited about our newly-accepted ICLR papers! Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theory (Spotlight π‘), SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agent (Spotlight π‘), and WebArena: A Realistic Web Environment for Building Autonomous Agents |
Nov 1, 2023 | New work on contextual privacy! ConfAIde: Can LLMs Keep a Secret? Testing Privacy Implications of Language Models! π Website π Tweet |
Oct 19, 2023 | New work on social intelligence! SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agents! π Website π Tweet |
Oct 16, 2023 | Really excited about our newly-accepted EMNLP papers:! Donβt Take This Out of Context! On the Need for Contextual Models and Evaluations for Stylistic Rewriting, FANTOM: A Benchmark for Analyzing Theory of Mind in Conversations, and our MRL workshop paper: Learning to translate by learning to communicate |
Jul 26, 2023 | New work on website navigation! WebArena: A Realistic Web Environment for Building Autonomous Agents! π Website ποΈ News Cover π Tweet |
Jun 16, 2023 | New preprint: Clever Hans or Neural Theory of Mind? Stress Testing Social Reasoning in Large Language Models! |
Jun 16, 2023 | New preprint: βDonβt Take This Out of Context!β On the Need for Contextual Models and Evaluations for Stylistic Rewriting! |
May 16, 2023 | Our paper COBRA π Frames: Contextual Reasoning about Effects and Harms of Offensive Statements appears at Findings of ACL 2023! |
Apr 16, 2023 | Our ToM Workshop has been accepted to ICML 2023! |
Apr 2, 2023 | Our team won the second place in the CMU LTI LLM hackathon! We focus on inversing the ChatGPT workflow! |
Jan 11, 2021 | Our paper: Challenges in Automated Debiasing for Toxic Language Detection will appear at EACL 2021! |
Dec 14, 2020 | Accept two intern offers from Apple! One happens at the Siri Info Intel team for the spring quarter, another happens at the Machine Translation team for the summer quarter |
Oct 20, 2020 | Our paper: Linguistically-Informed Transformations (LIT): A Method for Automatically Generating Contrast Sets will appear at BlackboxNLP 2020. |
Oct 2, 2020 | Our paper: Multilevel Text Alignment with Cross-Document Attention will appear at EMNLP 2020. |
Apr 20, 2020 | My undergraduate thesis: RPD: A Distance Function Between Word Embeddings get accepted by ACL SRW 2020. |