@davidweichiang

Associate Professor of Computer Science at University of Notre Dame. Artificial intelligence, natural language processing, formal language theory.

 South Bend, IN
nd.edu/~dchiang
Joined September 2012
  • Tweets817
  • Following592
  • Followers1.9K
36Media
David Chiang retweeted
Notre Dame CSE@ND_CSEOct 22
The Department of Computer Science & Engineering at @NotreDame is hiring tenure-track faculty (Assistant, Associate, or Full Professor) in Ethics & Computing 💻☘️ 📍 Notre Dame, IN 🗓️ Priority Deadline: Nov 15 🔗 Apply here: apply.interfolio.com/195819
3
6
355
David Chiang retweeted
TTIC@TTIC_ConnectOct 1
Congratulations to Professor Will Merrill (@lambdaviking), who received a Coefficient Giving grant to support his project, "From Transformers to Logic, Provably," to decode AI reasoning. Read more about his project: buff.ly/p4n1dT9
5
27
1,197
Et tu, Piggie?
1
8
150
David Chiang retweeted
Andy J Yang@pentagonalizeAug 14
Which state-tracking problems can transformers length generalize on? Transformers' performance on different finite-state tasks varies drastically, even when the tasks are structurally similar. Now, for the first time, we can precisely predict when length generalization occurs.
7
67
8
473
64,181
I am once again asking, why does @ReviewAcl make authors fill out the author registration form every single cycle?
2
3
62
6,049
Why does the form contain information that could be automatically obtained from the author's OpenReview profile?
1
11
502
And then why wonder that there is a reviewing crisis? It wastes too much of our time.
1
6
449
David Chiang retweeted
Andy J Yang@pentagonalizeJul 6
This is joint work with Pascal Bergsträßer, @GeorgZetzsche, @davidweichiang, and @anthonywlin to be presented at ICML 2026. Thu, Jul 9, 2026, 2:30 PM – 4:15 PM HALL A #1011 Looking forward to seeing everyone there!
1
3
256
Many algorithms (e.g., Turing machine simulations) have been written as transformers using average hard attention. Do these algorithms work with softmax attention? Please visit @pentagonalize, Dana Angluin, and my poster. Sunday 2pm Harbor H-I! aclanthology.org/2026.tacl-1…

Simulating Hard Attention Using Soft Attention

Andy Yang, Lena Strobl, David Chiang, Dana Angluin. Transactions of the Association for Computational Linguistics, Volume 14. 2026.

aclanthology.org
2
7
24
2,270
Correction: It will be a talk, not a poster. Sunday 3:10-3:20 in Harbor H-I!
1
4
229
Correction to my correction: contrary to the handbook, it will be the first in the session, so 2:00
1
131
*The Transformer Cookbook* by @pentagonalize @davidweichiang et al. A beautiful introduction to "hardcoding" algorithms inside the weights of a transformer (addition, lookup, branching) following the seminal RASP paper (srush.github.io/raspy/). openreview.net/forum?id=sPsh…
2
46
3
430
21,987
Replying to @ReviewAcl
@ReviewAcl Why does every author have to fill out the registration form if they already have an OpenReview profile?
2
3
1
24
2,518
And why does the registration form ask for so much information that is already in the OpenReview profile?
1
10
362
Congratulations! Everyone go work with @kentonmurray!
8
2,233
Most sentence embeddings capture *what* is said, not *how* it is said. We introduce ✨IDIOLEX ✨: a training framework that captures idiolect using only weak supervision (social proximity + linguistic features), not task labels. 📄 Preprint: arxiv.org/abs/2604.04704 🧵 1/8
2
12
1
44
9,877
David Chiang retweeted
Andy J Yang@pentagonalizeFeb 12
The FLaNN Workshop submission deadline has been extended to Feb 19! Invited talks + posters (non-archival): expressivity, computation, and learning in neural nets/LLMs. Previous work welcome. Graduate students encouraged to submit! 📍 Yale University 🗓️ May 11-13, 2026
Andy J Yang@pentagonalizeFeb 4
📣 FLaNN 2026 at Yale 🍮 Invited talks+posters (non-archival): expressivity, computation, and learning in neural nets/LLMs Speakers: Pablo Barceló, David Chiang, Will Merrill, Naomi Saphra, Gail Weiss Abstracts due Feb 12, 2026 Details: flann.cs.yale.edu/
1
7
10
1,591
David Chiang retweeted
Andy J Yang@pentagonalizeJan 30
Inviting submissions to the first Workshop on Formal Languages and Neural Networks! We welcome posters dicussing the formal expressivity, computational properties, and learning behavior of neural networks! Call for posters: flann.cs.yale.edu/cfp.html Deadline: February 12, 2026
Andy J Yang@pentagonalize19 Dec 2025
Announcing the first Workshop on Formal Languages and Neural Networks (FLaNN) 🍮! We invite the submission of abstracts for posters that discuss the formal expressivity, computational properties, and learning behavior of neural network models, including large language models.
2
11
2
21
6,819
Congratulations to @taguchi_c for finishing 2nd overall and 1st for the best improvement in any single language in the Mozilla Common Voice Spontaneous ASR shared task! codabench.org/competitions/1…
5
548
Announcing the first Workshop on Formal Languages and Neural Networks (FLaNN) 🍮! We invite the submission of abstracts for posters that discuss the formal expressivity, computational properties, and learning behavior of neural network models, including large language models.
1
14
4
34
11,726
#johdpapers New #datapaper out! 📰 "From Paginā to Webpage: On Developing and Documenting a Digitized Latin Collection" ✍️ Stephen Bothwell, Kaitlin Stephan, @gundimueller @davidweichiang ➡️ doi.org/10.5334/johd.397 #Latin #Digitization #TEI #OCR
2
5
387
Load more