Beiduo Chen

Our paper: arxiv.org/pdf/2505.23368
Our code: github.com/mainlp/CoT2EL

Thank you to my wonderful co-authors,
@janetlauyeung.bsky.social, Anna Korhonen, and @barbaraplank.bsky.social. Also to @mainlp.bsky.social , @cislmu.bsky.social @munichcenterml.bsky.social

See you in Suzhou!
#NLP #EMNLP2025

24.10.2025 13:42 — 👍 3    🔁 0    💬 0    📌 0

Matching exact probabilities for HLV is unstable. So, we propose a more robust rank-based evaluation that checks preference order. Our combined method outperforms baselines on 3 datasets that exhibit human label variation, showing it better aligns with diverse human perspectives.

24.10.2025 13:37 — 👍 0    🔁 0    💬 1    📌 0

Instead of unnatural post-hoc explanations, we look forward. A model's CoT already contains rationales for all options. We introduce CoT2EL, a pipeline that uses linguistic discourse segmenters to extract these high-quality, faithful units to explore human label variation.

24.10.2025 13:37 — 👍 0    🔁 0    💬 1    📌 0

📑 Our CoT2EL paper will be presented as an oral at #EMNLP2025 in Suzhou!

Humans often disagree on labels. Can a model's own reasoning (CoT) help us understand why? We developed a new method to extract these insights. Come join us!

🗓️ Friday, Nov 7, 14:00 - 15:30
📍 Room: A110

24.10.2025 13:36 — 👍 5    🔁 1    💬 1    📌 0

GitHub - mainlp/MJD-Estimator: Implementation of the EMNLP 2024 paper - "Seeing the Big through the Small": Can LLMs Approximate Human Judgment Distributions on NLI from a Few Explanations?; and the A... Implementation of the EMNLP 2024 paper - "Seeing the Big through the Small": Can LLMs Approximate Human Judgment Distributions on NLI from a Few Explanations?; and the ACL 2025 paper - A ...

🌍 Broader impact:
Our approach makes capturing disagreement scalable, helping build datasets that reflect real-world ambiguity—without requiring tons of human-written explanations.
Open-sourcing:
📂 github.com/mainlp/MJD-E...

15.07.2025 14:51 — 👍 3    🔁 0    💬 0    📌 0

🧠 What’s this about?
Human annotations often disagree. Instead of collapsing disagreement into a single label, we model Human Judgment Distributions — how likely humans are to choose each label in NLI tasks.
Capturing this is crucial for interpretability and uncertainty in NLP.

15.07.2025 14:50 — 👍 1    🔁 0    💬 0    📌 0

A Rose by Any Other Name: LLM-Generated Explanations Are Good Proxies for Human Explanations to Collect Label Distributions on NLI Disagreement in human labeling is ubiquitous, and can be captured in human judgment distributions (HJDs). Recent research has shown that explanations provide valuable information for understanding hum...

🔗 Paper link: arxiv.org/abs/2412.13942
🙏 Huge thanks to our collaborators Logan Siyao Peng, @barbaraplank.bsky.social, Anna Korhonen from @mainlp.bsky.social, @lmumuenchen.bsky.social, ‪@cambridgeltl.bsky.social ‬

15.07.2025 14:50 — 👍 2    🔁 0    💬 0    📌 0

🚨 Can LLMs generate explanations that are as useful as human ones for modeling label distributions in NLI?🌹"A Rose by Any Other Name" shows that they can
💬 We explore scalable, explanation-based annotation via LLMs.
📍Come find us in Vienna 🇦🇹! (July 28, 18:00-19:30, Hall 4/5) #ACL2025NLP #acl2025

15.07.2025 14:46 — 👍 5    🔁 1    💬 3    📌 0

The hand-drawn sign from three years ago.

🎉MaiNLP is turning 3 today!🎂🥳 We’ve grown a lot since @barbaraplank.bsky.social started this group with nothing but three aspiring researches and a hand-drawn sign on the door. Huge thanks to all the amazing people who have joined or visited us since. Here’s to many more years of exciting research!🚀

01.04.2025 10:40 — 👍 19    🔁 9    💬 1    📌 2