iiinova — where your digital self keeps evolving

Carolyn Anderson @carolynanderson.bsky.social · 21/05/2025

I'm excited about all of the fascinating work that will be presented this year at SCiL!

Reposted by Carolyn Anderson

Society for Computation in Linguistics @scilconference.bsky.social · 21/05/2025

SCiL 2025 is doing a second call for computational pragmatics methodology lightning talks as part of the special Symposium on Computational Pragmatics! You can submit a 100 word abstract until June 15th using this link: forms.gle/WRKvwcgjyHx3...

forms.gle

SCiL 2025 Pragmatics Methodology Lightning Talks

As part of the Symposium on Computational Pragmatics, we will hold a special "show-and-tell" session for research methodologies in computational pragmatics. We welcome lightning talks (no more than 5 ...

001

Carolyn Anderson @carolynanderson.bsky.social · 04/02/2025

Our new reasoning benchmark based on the NPR Sunday Puzzle's weekly challenge shows that o1 / o3-mini-high are significantly better at verbal reasoning than other models (i.e. R1)-- more below!

040

Reposted by Carolyn Anderson

Society for Computation in Linguistics @scilconference.bsky.social · 31/01/2025

Today is the last day to submit to SCiL 2025 (deadline is midnight anywhere on earth)! We look forward to your submissions! Submission instructions can be found here: wellesley-easel-lab.github.io/SCiL2025/ind...

wellesley-easel-lab.github.io

SCiL 2025

001

Reposted by Carolyn Anderson

Society for Computation in Linguistics @scilconference.bsky.social · 23/01/2025

8 more days to submit to SCiL 2025! If you have a neat modeling technique, experimental design, or other research method for computational pragmatics, please share it with the community by submitting a lightning talk! It doesn't have to be novel, just useful!

001

Carolyn Anderson @carolynanderson.bsky.social · 23/01/2025

excited to say that our Substance Beats Style paper was accepted to NAACL! We investigate *why* student-written programming prompts don't work well for LLMs, and find that while students think it's because of technical vocabulary gaps, it's actually information content that matters

arxiv.org

Substance Beats Style: Why Beginning Students Fail to Code with LLMs

Although LLMs are increasing the productivity of professional programmers, existing work shows that beginners struggle to prompt LLMs to solve text-to-code tasks. Why is this the case? This paper expl...

3348

Carolyn Anderson @carolynanderson.bsky.social · 16/01/2025

There are now two weeks to finish your SCiL 2025 submission! Especially excited to see work submitted to the Special Symposium on Computational Pragmatics!

043

Reposted by Carolyn Anderson

Society for Computation in Linguistics @scilconference.bsky.social · 14/01/2025

Ten more days to submit to SCiL 2025! We accept the following kinds of submissions: + Eight page papers or two-page abstracts on original research + Abstracts on work previously presented at a venue with a distinct scope and audience from SCiL + Methodology lightning talk proposals

wellesley-easel-lab.github.io

Call for Papers

001

Reposted by Carolyn Anderson

Society for Computation in Linguistics @scilconference.bsky.social · 08/01/2025

Did you know that SCiL accepts both abstracts (2 pages) and full papers (8 pages)? Both submission types will be considered for talks and posters. This year, we are also looking for lightning talks on methods in computational pragmatics!

021

Carolyn Anderson @carolynanderson.bsky.social · 07/01/2025

There's still plenty of time to put together a submission to SCiL 2025!

020

Reposted by Carolyn Anderson

Society for Computation in Linguistics @scilconference.bsky.social · 07/01/2025

The deadline for SCiL 2025 is in just a few weeks (Jan. 24th)! SCiL welcomes submissions on computational and mathematical approaches in any area of linguistics. SCiL 2025 will also include a special symposium on computational pragmatics! See the Call for Papers for more information:

wellesley-easel-lab.github.io

Call for Papers

033

Carolyn Anderson @carolynanderson.bsky.social · 04/01/2024

excited to be going to the LSA for the first time in a couple of years!

040

Carolyn Anderson @carolynanderson.bsky.social · 18/08/2023

Excited to release MultiPL-T, our method for generating semi-synthetic training data for low-resource programming languages! We achieve SOTA performance for Racket, OCaml, and Lua using a dataset of unit test validated translations of Python programs

khoury.northeastern.edu

Knowledge Transfer from High-Resource to Low-Resource Programming Languages for Code LLMs

180