CausalNLP is a research direction aiming to connect Causality and NLP #Causality #NLProc
Joined December 2021
- Tweets 18
- Following 13
- Followers 279
- Likes 7
4 Photos and videos
May 13
Sharing "Neuropathic Pain Diagnosis Simulator for Causal Discovery Algorithm Evaluation", one of the most popular datasets used for evaluating causal discovery algorithms! 🩺🧠 @RuiboTu @kunkzhang
📄 arxiv.org/abs/1906.01732
💻github.com/TURuibo/Neuropath…
1/n
1
102
May 13
🧪 Testing the SOTA:
The authors show that even top algorithms (PC, FCI, GES) struggle with this level of clinical logic, highlighting the need for more robust discovery methods.
1
40
May 13
🚀 Flexibility:
Researchers can tune the simulator to test for:
✅Unmeasured Confounding
✅ Selection Bias
✅ Missing Data (MCAR, MAR, MNAR)
A vital sandbox for moving Causal AI from "toy problems" to real-world medical utility. 📊
192
Apr 25
Sharing ACL 2024 Best Paper Winner, "Causal Estimation of Memorisation Profiles"!
LMs can reproduce training data verbatim, but measuring this "causally" (what would happen if the model never saw the data?) is hard. This paper fills the gap. link: aclanthology.org/2024.acl-lo…
1/n
1
1
3
525
Apr 25
⚙️ Training Dynamics:
Learning Rate (LR): Memorisation is heavily driven by the LR schedule. Instances seen during peak LR are memorised much more intensely. 📉
Order: Data order matters, but its effect is largely mediated by the LR and the specific model’s capacity. 🔀
1
41
Apr 25
🚀 Why it matters: Essential for understanding training dynamics, preventing copyright infringement, and enhancing data protection. 🛡️ By observing a small set of instances, we can now map a model's full "Memorisation Profile."
Experiments conducted on the Pythia model suite. 🧪
34
Apr 7
Introducing the latest research highlights from @JinesisLab at the Dagstuhl seminar on Causal LLMs!
1️⃣ CauSciBench: Evaluating LLM Causal Inference for Scientific Research
2️⃣ Causal AI Scientist: Facilitating Causal Data Science with LLMs
🔗 cr-llm.github.io/
1/n
1
2
9
1,250
Apr 7
1️⃣ CauSciBench: a benchmark assessing whether LLM agents 🤖 can answer scientific causal questions by navigating an end-to-end causal inference pipeline.
🔗 drive.google.com/file/d/17is…
1
1
70
Apr 7
2️⃣ Causal AI Scientist: an autonomous agent that performs end-to-end causal inference from a dataset, its description, and a causal query.
🔗 drive.google.com/file/d/1Kvx…
1
59
CausalNLP retweeted
Apr 7
We are hosting a Dagstuhl seminar on Causality & LLMs this week (Apr 7–10). Bringing together world experts to explore:
1️⃣ Integrating LLMs 🤖 into causal workflows
2️⃣ Evaluating & improving LLMs’ causal reasoning 🧠
Co-organized w/ @amt_shrma @DominikJanzing @kunkzhang @ZhijingJin
📍Schloss Dagstuhl, Wadern, Germany
🔗 dagstuhl.de/26152
📖 cr-llm.github.io/
📅 Apr 7–10
#CausalNLP #LLM #Dagstuhl @CausalNLP @MPI_IS @ELLISforEurope @UofTCompSci @VectorInst @TorontoSRI @CIFAR_News @JinesisLab @EuroSafeAI @ELLISInst_Tue
Also joined with my student @rahulbshrestha to present our CauSciBench and Causal AI Scientist work :)!
2
5
38
3,067
22 Dec 2022
Our CausalNLP Tutorial🎙️video and 🎞️slides are out! #EMNLP2022 @emnlpmeeting
Curious about (1) What Is Causality? and (2) How Causality Can Help #NLProc? Check out our #EMNLP2022 Tutorial with @ZhijingJin @amir_feder @kunkzhang
📺Video: youtu.be/4bq1ZYxXbtg
7
39
3,120
30 Oct 2022
Curious about "What is Causality & How can it help NLP"? Vote for your top interest areas to hear more about it during our #emnlp2022 Tutorial!
29 Oct 2022
We are preparing for the @CausalNLP Tutorial at #emnlp2022 @emnlpmeeting. 🙌 Welcome any "proposals" of topics you'd like to hear more about! Below are some candidates that we (@ZhijingJin @amir_feder @kunkzhang) are planning. We can talk more about the highly voted ones! 🎉
4
23 Dec 2021
A nice python-based causal discovery package by Prof Kun Zhang and his CMU team 👍
12 Nov 2021
We are excited to release the Python causal-learn package for causal discovery! See the package (github.com/cmu-phil/causal-l…) and documentation (causal-learn.readthedocs.io/…). Any feedback is welcome.
1