iiinova — where your digital self keeps evolving

Asher Zheng

@asher-zheng.bsky.social

171 followers 178 following 21 posts

PhD @ UT Linguistics Semantics/Pragmatics/NLP asherz720.github.io Prev.@UoEdinburgh @Hanyang

Posts Replies Media

Reposted by Asher Zheng

Ramya Namuduri @ramyanamuduri.bsky.social · 15/06/2026

Are LLM-generated stories novel? They can have unique characters and cliché plots, or the other way around. A holistic score doesn’t help distinguish the two 😔. Meet GENIE 🧞 – a fine-grained novelty metric that tells you where and why a response is original!

1134

Asher Zheng @asher-zheng.bsky.social · 13/06/2026

Spotting the rule from past experience is one thing; acting on it correctly is another. To find out, we introduce HERO's JOURNEY🦸‍♀️ to test for the LLMs’ inductive reasoning ability in multi-step setups. We found models show signs of rule induction, but scratch the surface.😮

1100

Reposted by Asher Zheng

Hongli Zhan @hongli-zhan.bsky.social · 29/04/2026

New paper! 🏁 Last one from my PhD at UT Austin. LLMs sound empathic but repeat the same discourse moves turn after turn — at 2x the rate of humans. We built MINT🌿, the first RL framework for discourse move diversity in empathic dialogue. +25% empathy, −26% repetition. 📄 arxiv.org/abs/2604.11742

1131

Asher Zheng @asher-zheng.bsky.social · 30/01/2026

Very excited to share that the paper w/ @jessyjli.bsky.social @DavidBeaver "Strategic Dialogue Assessment: The Crooked Path to Innocence" (used to have the name COBRA) was accepted by Dialogue and Discourse Vol 17 No.1. Check it out! 👉https://journals.uic.edu/ojs/index.php/dad/article/view/14503

252

Asher Zheng @asher-zheng.bsky.social · 03/06/2025

Language is often strategic, but LLMs tend to play nice. How strategic are they really? Probing into that is key for future safety alignment. 👉Introducing CoBRA🐍, a framework that assesses strategic language. Work with my amazing advisors @jessyjli.bsky.social and @David I. Beaver!

1102

Reposted by Asher Zheng

Ramya Namuduri @ramyanamuduri.bsky.social · 21/04/2025

Have that eerie feeling of déjà vu when reading model-generated text 👀, but can’t pinpoint the specific words or phrases 👀? ✨We introduce QUDsim, to quantify discourse similarities beyond lexical, syntactic, and content overlap.

2218