lexing xie

Democracy depends on an informed electorate. But political issues and ballot measures can be confusing, obscuring the effects of one outcome versus another. Moreover, politics is personal. Once we make an initial decision about an issue, it can be hard to change our mind or see things from “the other side.” And talking about issues with those with whom we disagree can be challenging, especially when the conversation feels more like a debate than a discussion. Technology offers ways to alleviate these difficulties, but not without introducing problems of its own. The Internet and social media promised new ways for people to connect, discuss issues, and learn from each other. But in practice, both often inflame passions, solidify echo chambers, and spread misinformation. More recently, LLM chat interfaces may help people stay informed through personalized access to information, but mainstream chatbots tend to match user beliefs rather than clarifying or challenging them.12 Without the kind of pushback you’d encounter in a discussion between disagreeing friends, chatbots are ill-suited for helping people think through political issues in a balanced way. The goal of CivicChats is to address these shortcomings. Starting with ballot measures, CivicChats helps people better understand political issues through three different modes of discussion: a Q&A mode for understanding what a measure does and what’s at stake, an argumentative mode that presents competing views to your own, and a reflective mode that helps you examine and develop your own thinking.

@emnlpmeeting (8/n) Read the paper here arxiv.org/abs/2509.24216 Explore the MoVa resources, data, and code supporting this work here: 👉 github.com/ZiyuChen0410/M… #NLP #ResearchTools #DataScience #compsocialsci #AIalignment

Identifying human morals and values in language is crucial for analysing lots of human- and AI-generated text. We introduce "MoVa: Towards Generalizable Classification of Human Morals and Values" - to be presented at @emnlpmeeting oral session next Thu 🧵 (1/n)

@emnlpmeeting (7/n) Led by CMlab PhD student Ziyu Chen, with @ChenhaoTan, Junfei Sun, Chenxi Li at UChicago -- thanks for inspiring all this during my 2024 sabbatical -- @joshnguyen99 at UPenn, and Jing Yao, Xiaoyuan Yi and Xing Xie at Microsoft Research Asia

@emnlpmeeting (6/n) Future work? Generalizable classification across cultures and languages, and investigating generalisable prompting methodolgy on other subjective text classification tasks.

@emnlpmeeting So what? MoVA could help: (1) Analyze Public Discourse: Understand the core values driving large-scale conversations on social and political issues. (2) Build Better AI: Ensure that artificial intelligence systems communicate in a way that's aligned with basic human ethics.

@emnlpmeeting (5/n) MoVa also offers a new application to evaluate psychological surveys: by scoring relevance of moral dimensions for each survey item, we can detect potentially multi-loaded items in instruments like MFQ, MAQ, and PVQ, helping researchers rethink questionnaire design.

@emnlpmeeting (4/n) MoVa provides resources defining this generalizable classification, including 16 labeled datasets and benchmarking results across four major, theoretically-grounded frameworks: Moral Foundations Theory (MFT), Human Values, Common Morality, and Morality-as-Cooperation (MAC)

@emnlpmeeting (3/n) Our new methodology insight: "all@once LLM prompting strategy" outperforms fine-tuned models across multiple domains and frameworks. Why does it work? It uses inter-label dependencies resembling a classifier chain approach in ML.

@emnlpmeeting (2/n) What is generalizable classification here? We think there are three key elements 1. New data domains - from short informal text to long passages. 2. New moral and value dimensions. 3. New frameworks - e.g. moral foundations, Schwartz human values, and many more!

The project will focus on complex information needs in climate change and other social purposes. We are open to new algorithms, paradigms for human-AI collaboration, and innovations with LLMs.

Announcing 2x Research Fellows positions on NLP and machine learning, working w wonderful Jing Jiang, Thang Bui and myself + partners in Data61 - apply by 31 Dec or help spread the words jobs.anu.edu.au/jobs/research-… more detail in thread

@hengjinlp @eagle_hz second this! We started working with Qingyun recently, already enjoying the knowledge, enthusiasm and collaborative spirit he brings.

@TonyWirthPhD congrats to being in a new big city and the start of a new adventure! hope to catch up sometime

First selfie yesterday… found a tree and a new name tag! #USydCompSci

@TingAstro Wonderful sights and touching adventure 👍

I discovered new cousins, family bonds unbroken by time or distance. And, standing in the village my grandfather yearned for, I finally understood the meaning behind my name. 4/4

My name, Yuan Sen, holds an echo of history. It means 'the source of the river in a forest,' a poetic tribute from my father to commemorate my grandfather's past. This was the man forced to leave his village behind for Borneo, carrying an unfulfilled yearning in his heart. 1/4

@haldaume3 This is just one of many ways to partition the content about limitation and broader implications. What other strategy have you used? (4/4)

@haldaume3 at the end of the paper, “for practitioners adopting this, it is worth considering additional metrics c, d, e, and be cautious of blah blah”, “we have used domain X as examples, the methods also apply to domain U and V, for domains Y and Z important variations include blah” (3/4)

I get the need for making papers include discussions of limitations - appropriately scoping work is a big problem in AI. I also really dislike writing "limitations sections" - good prose should make limitations clear when ideas and results are introduced, not as an afterthought.

@berty38 Hope you're ok Bert! Safe travels on the future paths you will take + I am confident that wherever you are, that corner of the world will become a bit better because of you.

I’m going to take some time to figure out what to do next professionally. I still hope to find a path where I can use my knowledge to make the world a better place—all while having the balance to take care of myself and do other things I love in addition to work. Bert 2/2

Dear friends, I’ve spent the past year on leave from work for personal health reasons. Now, with a lot of other life changes, I’ve finalized the decision to leave Tufts this summer.