Skip to content
AI News HubLIVE
Original source2 min read

You Really Didn't Get That? Benchmarking Social Pragmatic Inference for Indirect and Playful Chinese Online Comments

Summary

This paper introduces a benchmark for evaluating whether LLMs can recover situated pragmatic meanings from indirect and playful Chinese online comments. Built from over 200,000 public Chinese social media interaction records, the benchmark contains 4,735 human-validated diagnostic items. Across eight LLMs, the mean leave-writer-out accuracy is 68.70% and the best model reaches 81.42%, while human accuracy is 90.8%. Case analyses show models often detect broad irony or playfulness but fail to identify the precise mechanism or interactional move.

SourcearXiv Computational LinguisticsAuthor: Shiwei Hong, Junjie Ma, Emma Jiren Wang, Ethan Z. Rong, Siying Hu, Haichang Li, Ziying Wang, Zhicong Lu
You Really Didn't Get That? Benchmarking Social Pragmatic Inference for Indirect and Playful Chinese Online Comments
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 3 Sep 2026]

Title:You Really Didn't Get That? Benchmarking Social Pragmatic Inference for Indirect and Playful Chinese Online Comments

View a PDF of the paper titled You Really Didn't Get That? Benchmarking Social Pragmatic Inference for Indirect and Playful Chinese Online Comments, by Shiwei Hong and 7 other authors

View PDF HTML (experimental)

Abstract:Chinese online comments often convey social meaning through indirect and playful language that is hard to interpret without context. Existing evaluations largely organize items around predefined phenomena or controlled pragmatic categories, leaving open whether models can distinguish plausible readings of what a naturally occurring comment is doing in a particular exchange. We introduce a benchmark for evaluating whether LLMs can recover such situated pragmatic meanings. From more than 200,000 public Chinese social media interaction records, we construct 4,735 human-validated diagnostic items, each pairing a target comment with reconstructed preceding context and plausible misreadings. We evaluate eight LLMs as both question writers and solvers in a cross-writer setting. The task is challenging: the strongest model achieves 81.42% leave-writer-out accuracy. Across all eight models, the mean leave-writer-out accuracy is 68.70% while human accuracy was 90.8%. Case analysis shows that models often recognize broad irony or playfulness while misidentifying the mechanism or interactional move.

Comments: Accepted to the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP 2026), Main Conference

Subjects:

Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)

Cite as: arXiv:2609.04384 [cs.CL]

(or arXiv:2609.04384v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2609.04384

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Shiwei Hong [view email] [v1] Thu, 3 Sep 2026 18:45:54 UTC (207 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled You Really Didn't Get That? Benchmarking Social Pragmatic Inference for Indirect and Playful Chinese Online Comments, by Shiwei Hong and 7 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CL

new | recent | 2026-09

Change to browse by:

cs cs.AI cs.HC

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

InvestorsAdvanced

Key points

  • A new benchmark uses 4,735 human-validated items pairing Chinese comments with context and plausible misreadings.
  • Across eight LLMs, average leave-writer-out accuracy is 68.70%; the best model scores 81.42%, versus 90.8% for humans.
  • Models tend to recognize generic irony or playfulness yet misidentify the specific mechanism or interactional intent.
  • The paper was accepted to the EMNLP 2026 main conference (arXiv:2609.04384).

Highlights and analysis are generated automatically and may contain errors. Check the original source.

More recent stories on this topic · Models