There are so many scoping reviews recently, I'm patiently waiting for the first scoping review of scoping reviews

Since when has the Learn mode in Quizlet not been free! And are there any free alternatives? (ideally with games/gamification elements)

Desk rejected after 1 hour and 16 minutes...has anyone done better than that? 😅

Just out: Statistical Modeling for (Actual) Hypothesis Testing: Building Cumulative Knowledge in Corpus Linguistics (Larsson & Hancock, 2026) Download for free until August 11: doi.org/10.1017/9781... On methods for evaluating empirically informed hypotheses. @cambup-linguistics.cambridge.org

Statistical Modeling for (Actual) Hypothesis Testing

Cambridge Core - Research Methods in Linguistics - Statistical Modeling for (Actual) Hypothesis Testing

doi.org

Interview with Nick Ellis from about 7 years ago about his 2002 Frequency SSLA Article: youtu.be/LpKvMBPHV-k?...

An interview with Nick Ellis for Studies in Second Language Acquisition

YouTube video by Cambridge University Press

youtu.be

I wonder if there is much research out there on the metaphors AI uses, I seem to get a lot of "knobs" and recently "sidecars" from Claude

100 words is too short for an abstract!!!

I like the (relatively new?) Sketch Engine AI search for quick searches, e.g. "break collocates associated with the human body (e.g. arm, wrist, bone)" CQL: [ws("break-v", ".*", "(arm|wrist|bone|leg|hand|finger|toe|foot|neck|back|rib|nose|skull|ankle|elbow|hip|shoulder)-n")]

so many PDFs to read, so little time 🤓

👓 Have you encountered our new format 𝗔𝗦𝗣𝗘𝗖𝗧𝗦 yet? 💡 𝙈𝙤𝙧𝙚 𝙘𝙤𝙢𝙥𝙖𝙘𝙩 𝙩𝙝𝙖𝙣 𝙖 𝙢𝙤𝙣𝙤𝙜𝙧𝙖𝙥𝙝, 𝙡𝙚𝙨𝙨 𝙡𝙞𝙢𝙞𝙩𝙚𝙙 𝙩𝙝𝙖𝙣 𝙖 𝙧𝙚𝙜𝙪𝙡𝙖𝙧 𝙟𝙤𝙪𝙧𝙣𝙖𝙡 𝙖𝙧𝙩𝙞𝙘𝙡𝙚! Applied: www.benjamins.com/catalog/aspal Corpus: www.benjamins.com/catalog/aspcl Pragmatics: www.benjamins.com/catalog/aspp TS: www.benjamins.com/catalog/aspts

Butter consumption correlates with wind power generated in the United States 🫠... And other correlations www.tylervigen.com/spurious-cor...

Spurious Correlations

Correlation is not causation: thousands of charts of real data showing actual correlations between ridiculous variables.

tylervigen.com

Day 2 and got over the rejection rage🤬 enough to start on revisions☺️. Still think they could have given us major revisions instead of reject...

Rejected 😭. Shouldn't have checked my email just before the England game!

Moore, Caines & Buttery (2026) Multi-component student writing profiles for expert-aligned automated evaluation of English learner essays aclanthology.org/2026.bea-1.39/ This looks very interesting, using multiple drafts from learners in the Write & Improve Corpus

Multi-component student writing profiles for expert-aligned automated evaluation of English learner essays.

Russell Moore, Andrew Caines, Paula Buttery. Proceedings of the 21st Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2026). 2026.

aclanthology.org

Claude's always taking back "I should search past conversations to check whether I actually did help write this notebook, since I don't have direct memory of that and shouldn't just accept the framing uncritically" trust me you did+the part you added was wrong & needs correcting

Finally got my name on the door after almost 3 months

How do emojis & homoglyphs affect corpus tokenisation, and why does it matter? 🧐 OUT NOW: @matteodic.bsky.social explores how small discrepancies in tokenisation can distort linguistic analysis and undermine data fidelity 🔍 doi.org/10.1075/ijcl...

Bayesian teaching enables probabilistic reasoning in large language models (Qiu et al., 2026) www.nature.com/articles/s41...

Bayesian teaching enables probabilistic reasoning in large language models - Nature Communications

LLMs fail to update beliefs in a Bayesian way but can be taught to do so by mimicking a normative Bayesian model. This training yields better predictions and transferable reasoning skills across tasks...

nature.com

"I'll clone the repo directly in the sandbox", OK I'll leave you to it Claude...

Prof Yeh, AI by Hand "I see a few benefits. One is that I can only go at the humanly possible speed of my writing. I cannot go any faster than I can write. Second students can only learn at a humanly possible speed. They can only follow how much I write." youtu.be/BAgxGp2WEu4?...

A CS Professor on Why Slow Learning Wins in the AI Era | CU Boulder, Tom Yeh

YouTube video by EO

youtu.be

AI for smell... "For the first time, you can turn any word, image, or idea into an unforgettable, custom fragrance..." 🫢 www.osmo.ai

*awaiting reviewer assignment has reached the 3 month mark 🫣

Just trying a local LLM for the first time (Gemma 4 with LM Studio). For some reason running a local LLM on your own machine feels more like a Frankenstein than using one of the online chatbot LLMs 🧟‍♂️

Looks interesting, research utilising the English Grammar Profile and corpus query patterns Linking textbook grammar input to writing accuracy in exam-oriented EF... www.sciencedirect.com/science/arti...

Linking textbook grammar input to writing accuracy in exam-oriented EFL systems: A CEFR-B1, corpus-based study

In exam-oriented EFL systems, textbooks often function as the practical syllabus for high-stakes writing. This study examined whether corpus-derived p…

sciencedirect.com

Abstract submitted for The 6th Asia Pacific Corpus Linguistics Conference (APCLC2026) in Toyama smartconf.jp/content/apcl...

HOME

smartconf.jp

Researchers using corpus-based approaches to the study of English: Welcome to the ICAME48 conference in Helsinki in May 2027! Follow our BlueSky account @icame48.bsky.social and our website (see the link in comments) for updates. The CfP will soon be published. #ICAME48

My interview with Prof. Mizumoto has been re-released in the "legends series" of Lost in Citations www.podbean.com/ew/pb-kn3xm-... The tool we talked about was later developed into Auto Error Analyzer langtech.jp/autoerror/ A paper about it was published in SSLA www.cambridge.org/core/journal...

Why do links to the IRIS database in papers never seem to work, they just seem to go to the IRIS homepage 🤔

This is why you should make your own table and not just add the image of the output from the stats software isn't it? (completely unreadable)

Editors of semantics journal resign, launch new journal after publisher ‘ultimatum’ – Retraction Watch share.google/7l0FgC2wdQbW...

Editors of semantics journal resign, launch new journal after publisher ‘ultimatum’

The new journal, Semantics of Natural Languages, launched in May. The editors of a semantics journal owned by Springer Nature have resigned to launch a new one, citing pressure from the company to …

share.google