Workshop
Causality and Large Models
Felix Leeb · Ching Lam Choi · Luigi Gresele · Josef Valvoda · Andrei Nicolicioiu · Xiusi Li · Patrik Reizinger · Sophie Xhonneux · Haoxuan Li · Mengyue Yang · Bernhard Schölkopf · Dhanya Sridhar
East Exhibition Hall C
Sat 14 Dec, 8:45 a.m. PST
Our workshop aims to explore the synergies between causality and large models, also known as ``foundation models,'' which have demonstrated remarkable capabilities across multiple modalities (text, images, audio, etc.). Despite their high performance, the opaque nature of these large models raises crucial questions regarding their trustworthiness, especially in safety-critical domains. A growing community of researchers is turning towards a more principled framework to address these concerns, better understand the behavior of large models, and improve their reliability: causality.Specifically, this workshop will focus on four directions: causality in large models, to assess their causal reasoning abilities, causality for improving large models, causality with large models to enhance causal inference and discovery methods, and causality of large models to understand and control their internal mechanisms. The invited speakers and panelists (almost all of which have already been confirmed to attend) represent a diverse set of perspectives and expertise, across both academia and industry.The workshop is organized by a team of 12 members from six different institutions across North America, Europe, and Asia, ensuring diversity across research interests, backgrounds, and demographics. Visit our website: https://calm-workshop-2024.github.io/
Schedule
Sat 8:45 a.m. - 9:00 a.m.
|
Opening Remarks
(
Intro
)
>
SlidesLive Video |
🔗 |
Sat 9:00 a.m. - 9:30 a.m.
|
Teaching causal reasoning to language models
(
Invited Talk
)
>
SlidesLive Video |
Amit Sharma 🔗 |
Sat 9:30 a.m. - 10:00 a.m.
|
Causal reasoning in foundation agents
(
Invited Talk
)
>
SlidesLive Video |
Jane Wang 🔗 |
Sat 10:30 a.m. - 11:00 a.m.
|
Towards Causal Artificial Intelligence
(
Invited Talk
)
>
SlidesLive Video |
Elias Bareinboim 🔗 |
Sat 11:00 a.m. - 11:15 a.m.
|
From Causal to Concept-Based Representation Learning
(
Oral
)
>
link
SlidesLive Video |
Goutham Rajendran · Simon Buchholz · Bryon Aragam · Bernhard Schölkopf · Pradeep Ravikumar 🔗 |
Sat 11:15 a.m. - 11:30 a.m.
|
Causal Order: The Key to Leveraging Imperfect Experts in Causal Inference
(
Oral
)
>
link
SlidesLive Video |
Aniket Vashishtha · Abbavaram Gowtham Reddy · Abhinav Kumar · Saketh Bachu · Vineeth N Balasubramanian · Amit Sharma 🔗 |
Sat 11:30 a.m. - 12:00 p.m.
|
Poster Session 1
(
Poster Session
)
>
|
🔗 |
Sat 1:30 p.m. - 2:00 p.m.
|
Musings on the Linear Representation Hypothesis
(
Invited Talk
)
>
SlidesLive Video |
Victor Veitch 🔗 |
Sat 2:00 p.m. - 2:15 p.m.
|
Using Relational and Causality Context for Tasks with Specialized Vocabularies that are Challenging for LLMs
(
Oral
)
>
link
SlidesLive Video |
Ryosuke Nakanishi · Yan-Ying Chen · Francine Chen · Matt Klenk · Charlene C. Wu 🔗 |
Sat 2:15 p.m. - 2:30 p.m.
|
Causally Testing Gender Bias in LLMs: A Case Study on Occupational Bias
(
Oral
)
>
link
SlidesLive Video |
Yuen Chen · Vethavikashini Chithrra Raghuram · Justus Mattern · Rada Mihalcea · Zhijing Jin 🔗 |
Sat 2:30 p.m. - 3:00 p.m.
|
Poster Session 2
(
Poster Session
)
>
|
🔗 |
Sat 3:30 p.m. - 4:00 p.m.
|
Hypothesis testing the circuit hypothesis in LLMs
(
Invited Talk
)
>
SlidesLive Video |
Claudia shi 🔗 |
Sat 4:00 p.m. - 4:30 p.m.
|
Causal thinking in humans and machines
(
Invited Talk
)
>
SlidesLive Video |
Tobias Gerstenberg 🔗 |
Sat 4:30 p.m. - 5:30 p.m.
|
Panel Discussion
(
Panel Discussion
)
>
SlidesLive Video |
Atticus Geiger · Chelsea Finn · Zhijing Jin · Giambattista Parascandolo · Maria Antoniak · Elias Bareinboim 🔗 |
-
|
Causal Reasoning in Large Language Models: A Knowledge Graph Approach ( Poster ) > link | Yejin Kim · Eojin Kang · Juae Kim · H. Howie Huang 🔗 |
-
|
CausalGraph2LLM: Evaluating LLMs for Causal Queries ( Poster ) > link | Ivaxi Sheth · Bahare Fatemi · Mario Fritz 🔗 |
-
|
Teaching Transformers Causal Reasoning through Axiomatic Training ( Poster ) > link | Aniket Vashishtha · Abhinav Kumar · Atharva Pandey · Abbavaram Gowtham Reddy · Vineeth N Balasubramanian · Amit Sharma 🔗 |
-
|
Reasoning with a Few Good Cross-Questions Greatly Enhances Causal Event Attribution in LLMs ( Poster ) > link | Sanyam Saxena · Sunita Sarawagi 🔗 |
-
|
From Correlation to Causation: Understanding Climate Change through ML and LLM Inquiries ( Poster ) > link | Shan Shan 🔗 |
-
|
Using Relational and Causality Context for Tasks with Specialized Vocabularies that are Challenging for LLMs ( Poster ) > link | Ryosuke Nakanishi · Yan-Ying Chen · Francine Chen · Matt Klenk · Charlene C. Wu 🔗 |
-
|
Evaluating Interventional Reasoning Capabilities of Large Language Models ( Poster ) > link | Tejas Kasetty · Divyat Mahajan · Gintare Karolina Dziugaite · Alexandre Drouin · Dhanya Sridhar 🔗 |
-
|
Counterfactual Causal Inference in Natural Language with Large Language Models ( Poster ) > link | Gaël Gendron · Jože Rožanec · Michael Witbrock · Gillian Dobbie 🔗 |
-
|
On Incorporating Prior Knowledge Extracted from Pre-trained Language Models into Causal Discovery ( Poster ) > link |
14 presentersChanhui Lee · Juhyeon Kim · YongJun Jeong · Yoonseok Yeom · Juhyun Lyu · Jung-Hee Kim · Sangmin Lee · Sangjun Han · Hyeokjun Choe · Soyeon Park · Woohyung Lim · Kyunghoon Bae · Sungbin Lim · Sanghack Lee |
-
|
Hypothesizing Missing Causal Variables with LLMs ( Poster ) > link | Ivaxi Sheth · Sahar Abdelnabi · Mario Fritz 🔗 |
-
|
CausalBench: A Comprehensive Benchmark for Evaluating Causal Reasoning Capabilities of Large Language Models ( Poster ) > link | ZEYU WANG 🔗 |
-
|
Competence-Based Analysis of Language Models ( Poster ) > link | Adam Davies · Jize Jiang · Cheng Xiang Zhai 🔗 |
-
|
On LLM Augmented AB Experimentation ( Poster ) > link | Shiv Shankar · Ritwik Sinha · Madalina Fiterau 🔗 |
-
|
Can large language models reason about causal relationships in multimodal time series data? ( Poster ) > link | Elizabeth Healey · Isaac S Kohane 🔗 |
-
|
Are UFOs Driving Innovation? The Illusion of Causality in Large Language Models ( Poster ) > link | Maria Vcitoria Carro · Francisca Selasco · Denise Alejandra Mester · Mario Leiva 🔗 |
-
|
Interactive Semantic Interventions for VLMs: A Causality-Inspired Investigation of VLM Failures ( Poster ) > link | Lukas Klein · Kenza Amara · Carsten Lüth · Hendrik Strobelt · Mennatallah El-Assady · Paul Jaeger 🔗 |
-
|
Estimating Effects of Tokens in Preference Learning ( Poster ) > link | Hsiao-Ru Pan · Maximilian Mordig · Bernhard Schölkopf 🔗 |
-
|
Causally Testing Gender Bias in LLMs: A Case Study on Occupational Bias ( Poster ) > link | Yuen Chen · Vethavikashini Chithrra Raghuram · Justus Mattern · Rada Mihalcea · Zhijing Jin 🔗 |
-
|
Counterfactual Token Generation in Large Language Models ( Poster ) > link | Ivi Chatzi · Nina Corvelo Benz · Eleni Straitouri · Stratis Tsirtsis · Manuel Rodriguez 🔗 |
-
|
Causal World Representation in the GPT Model ( Poster ) > link | Raanan Rohekar · Yaniv Gurwicz · Sungduk Yu · VASUDEV LAL 🔗 |
-
|
CausalQuest: Collecting Natural Causal Questions for AI Agents ( Poster ) > link | Roberto Ceraolo · Dmitrii Kharlapenko · Amélie Reymond · Rada Mihalcea · Bernhard Schölkopf · Mrinmaya Sachan · Zhijing Jin 🔗 |
-
|
A Causal Perspective in Brainwave Foundation Models ( Poster ) > link | Konstantinos Barmpas · Yannis Panagakis · Dimitrios Adamos · N Laskaris · Stefanos Zafeiriou 🔗 |
-
|
From Causal to Concept-Based Representation Learning ( Poster ) > link | Goutham Rajendran · Simon Buchholz · Bryon Aragam · Bernhard Schölkopf · Pradeep Ravikumar 🔗 |
-
|
Investigating Causal Reasoning in Large Language Models ( Poster ) > link | Atul Rawal · Raglin · Qianlong Wang · Ziying Tang 🔗 |
-
|
Leveraging LLM-Generated Structural Prior for Causal Inference with Concurrent Causes ( Poster ) > link | Xingjian Zhang · Shixuan Liu · Yixin Wang · Qiaozhu Mei 🔗 |
-
|
Causal Interventions on Causal Paths: Mapping GPT-2's Reasoning From Syntax to Semantics ( Poster ) > link | Isabelle Lee · Joshua Lum · Ziyi Liu · Dani Yogatama 🔗 |
-
|
CodeSCM: Causal Analysis for Multi-Modal Code Generation ( Poster ) > link | Mukur Gupta · Noopur Bhatt · Suman Jana 🔗 |
-
|
Are Police Biased? An NLP Approach ( Poster ) > link | Jonathan Choi 🔗 |
-
|
Causal Order: The Key to Leveraging Imperfect Experts in Causal Inference ( Poster ) > link | Aniket Vashishtha · Abbavaram Gowtham Reddy · Abhinav Kumar · Saketh Bachu · Vineeth N Balasubramanian · Amit Sharma 🔗 |
-
|
LLM-initialized Differentiable Causal Discovery ( Poster ) > link | Shiv Kampani · David Hidary · Constantijn van der Poel · Martin Ganahl · Brenda Miao 🔗 |