Merge branch 'master' of https://github.com/cicl-stanford/homepage

content/home/people.md

@@ -26,15 +26,6 @@ weight = 3
 	website = "tobias_gerstenberg"
 	description = "I am interested in how people hold others responsible, how these judgments are grounded in causal representations of the world, and supported by counterfactual simulations. I also like to drink tea."
 
-[[member]]
-	id = "Lara Kirfel"
-	position = "Postdoctoral Researcher"
-	email = "[email protected]"
-	github = "LaraKirfel"
-    scholar = "citations?user=uJkMLZwAAAAJ&hl=de"
-    website = "https://larakirfel.tumblr.com/"
-    description = "My research focuses on causal reasoning, moral cognition and counterfactual thinking with the occasional dip into philosophy. I like to drink strong black coffee."
-
 [[member]]
 	id = "Jan-Philipp Fränken"
 	position = "Postdoctoral Researcher"
@@ -140,15 +131,6 @@ weight = 3
     email = "[email protected]"
     description = "I'm passionate with complex systems, human intellectuality, and world-building prospects. Majoring in computer science, I’m excited about AI and interdisciplinary research. Things I enjoy: poetry and guitar, hiking and dancing, and a sip of Earl Grey Tea."
 
-[[member]]
-    id = "Haoran Zhao"
-    position = "Research Assistant"
-    email = "[email protected]"
-    github = "haoranzhao419"
-    twitter = "HaoranZhaoHRZ"
-    website = "https://haoranzhao419.github.io/"
-    description = "I am interested in understanding how effectively language models can perform reasoning and comprehend commonsense and factual knowledge. Subsequently, with a better understanding of LLM’s cognitive abilities, I hope to build more cognitive-feasible and efficient language models at small scales. In my free time, I like running and sailing. I like lemonade."
-
 [[member]]
     id = "Sunny Yu"
     position = "Research Assistant"
@@ -508,8 +490,26 @@ weight = 3
 #     github = "ggaw25"
 #     description = "I’m interested in understanding how people make complex decisions through computational models. I am currently a sophomore studying Computer Science and Economics. I love all things sports and outdoors, and light-roasted black coffee."
 
+# [[member]]
+# id = "Lara Kirfel"
+# position = "Postdoctoral Researcher"
+# email = "[email protected]"
+# github = "LaraKirfel"
+# scholar = "citations?user=uJkMLZwAAAAJ&hl=de"
+# website = "https://larakirfel.tumblr.com/"
+# description = "My research focuses on causal reasoning, moral cognition and counterfactual thinking with the occasional dip into philosophy. I like to drink # strong black coffee."
+
+# [[member]]
+# id = "Haoran Zhao"
+# position = "Research Assistant"
+# email = "[email protected]"
+# github = "haoranzhao419"
+# twitter = "HaoranZhaoHRZ"
+# website = "https://haoranzhao419.github.io/"
+# description = "I am interested in understanding how effectively language models can perform reasoning and comprehend commonsense and factual knowledge. Subsequently, with a better understanding of LLM’s cognitive abilities, I hope to build more cognitive-feasible and efficient language models at small scales. In my free time, I like running and sailing. I like lemonade."
+
 [[member]]
 id = "Alumni"
-description = "<ul><li> <a href='https://www.cmu.edu/dietrich/philosophy/people/masters/damini-kusum.html'>Damini Kusum</a> (research assistant): Now MSc student at Carnegie Mellon University. <li> <a href='https://josephouta.com/'>Joseph Outa</a> (research assistant): Now PhD student at Johns Hopkins University. </li><li> <a href='https://zach-davis.github.io/'>Zach Davis</a> (postdoc): Now research scientist at Facebook Reality Labs. </li> <li><a href='https://www.linkedin.com/in/erin-bennett-a1a9623a'>Erin Bennett</a> (lab affiliate)</li> <li>Bryce Linford (research assistant): Now PhD student at UCLA.</li> <li><a href='https://scholar.google.com/citations?user=R2Ji5Z8AAAAJ&hl=en'>Antonia Langenhoff</a> (research assistant): Now PhD student at UC Berkeley.</li> </ul>"
+description = "<ul><li> <a href='https://www.mpib-berlin.mpg.de/person/lara-kirfel/367762'>Lara Kirfel</a> (postdoc): Now Postdoctoral Fellow at the Center for Humans and Machines, MPI Berlin.<li> <a href='https://www.cmu.edu/dietrich/philosophy/people/masters/damini-kusum.html'>Damini Kusum</a> (research assistant): Now MSc student at Carnegie Mellon University. <li> <a href='https://josephouta.com/'>Joseph Outa</a> (research assistant): Now PhD student at Johns Hopkins University. </li><li> <a href='https://zach-davis.github.io/'>Zach Davis</a> (postdoc): Now research scientist at Facebook Reality Labs. </li> <li><a href='https://www.linkedin.com/in/erin-bennett-a1a9623a'>Erin Bennett</a> (lab affiliate)</li> <li>Bryce Linford (research assistant): Now PhD student at UCLA.</li> <li><a href='https://scholar.google.com/citations?user=R2Ji5Z8AAAAJ&hl=en'>Antonia Langenhoff</a> (research assistant): Now PhD student at UC Berkeley.</li> </ul>"
 
 +++

content/publication/amemiya2024disagreement.md

@@ -0,0 +1,33 @@
++++
+# 0 -> 'Forthcoming',
+# 1 -> 'Preprint',
+# 2 -> 'Journal',
+# 3 -> 'Conference Proceedings',
+# 4 -> 'Book chapter',
+# 5 -> 'Thesis'
+
+title = "Children use disagreement to infer what happened"
+date = "2024-05-13"
+authors = ["J. Amemiya","G. D. Heyman","T. Gerstenberg"]
+publication_types = ["2"]
+publication_short = "_Cognition_"
+publication = "Amemiya J., Heyman G. D., Gerstenberg T. (2024). Children use disagreement to infer what happened. _Cognition_."
+abstract = "In a rapidly changing and diverse world, the ability to reason about conflicting perspectives is critical for effective communication, collaboration, and critical thinking. The current pre-registered experiments with children ages 7 to 11 years investigated the developmental foundations of this ability through a novel social reasoning paradigm and a computational approach. In the inference task, children were asked to figure out what happened based on whether two speakers agreed or disagreed in their interpretation. In the prediction task, children were provided information about what happened and asked to predict whether two speakers will agree or disagree. Together, these experiments assessed children's understanding that disagreement often results from ambiguity about what happened, and that ambiguity about what happened is often predictive of disagreement. Experiment 1 (N = 52) showed that children are more likely to infer that an ambiguous utterance occurred after learning that people disagreed (versus agreed) about what happened and found that these inferences become stronger with age. Experiment 2 (N = 110) similarly found age-related change in children's inferences and also showed that children could reason in the forward direction, predicting that an ambiguous utterance would lead to disagreement. A computational model indicated that although children's ability to predict when disagreements might arise may be critical for making the reverse inferences, it did not fully account for age-related change."
+image_preview = ""
+selected = false
+projects = []
+url_pdf = "papers/amemiya2024disagreement.pdf"
+url_preprint = "https://psyarxiv.com/y79sd/"
+url_code = ""
+url_dataset = ""
+url_slides = ""
+url_video = ""
+url_poster = "posters/amemiya2023disagreement-poster.pdf"
+url_source = ""
+url_custom = [{name = "Github", url = "https://github.com/cicl-stanford/children_disagree"}]
+math = true
+highlight = true
+[header]
+# image = "publications/amemiya2023disagreement.png"
+caption = ""
++++

content/publication/brockbank2024monster.md

@@ -0,0 +1,33 @@
++++
+# 0 -> 'Forthcoming',
+# 1 -> 'Preprint',
+# 2 -> 'Journal',
+# 3 -> 'Conference Proceedings',
+# 4 -> 'Book chapter',
+# 5 -> 'Thesis'
+
+title = "Without his cookies, he's just a monster: A counterfactual simulation model of social explanation"
+date = "2024-05-12"
+authors = ["E. Brockbank", "J. Yang", "M. Govil","J. E. Fan","T. Gerstenberg"]
+publication_types = ["3"]
+publication_short = "_Proceedings of the 46th Annual Conference of the Cognitive Science Society_"
+publication = "Brockbank, E., Yang, J., Govil, M., Fan, J. E., Gerstenberg, T. (2024). Without his cookies, he's just a monster: A counterfactual simulation model of social explanation. In _Proceedings of the 46th Annual Conference of the Cognitive Science Society_."
+abstract = "Everyday reasoning about others involves accounting for why they act the way they do. With many explanations for someone's behavior, how do observers choose the best one? A large body of work in social psychology suggests that people's explanations rely heavily on traits rather than external factors. Recent results have called this into question, arguing that people balance traits, mental states, and situation to make sense of others' actions. How might they achieve this? In the current work, we hypothesize that people rely on counterfactual simulation to weigh different explanations for others' behavior. We propose a computational model of this process that makes concrete predictions about when people will prefer to explain events based on the actor's traits or their situation. We test the predictions of this model in an experimental paradigm in which trait and situation each guide behavior to varying degrees. Our model predicts people's causal judgments well overall but is less accurate for trait explanations than situational explanations. In a comparison with simpler causal heuristics, a majority of participants were better predicted by the counterfactual model. These results point the way toward a more comprehensive understanding of how social reasoning is performed within the context of domain-general causal inference."
+image_preview = ""
+selected = false
+projects = []
+url_pdf = "papers/brockbank2024monster.pdf"
+url_preprint = "https://osf.io/preprints/psyarxiv/3wzbk"
+url_code = ""
+url_dataset = ""
+url_slides = ""
+url_video = ""
+url_poster = ""
+url_source = ""
+url_custom = [{name = "Github", url = "https://github.com/cicl-stanford/action_abstraction_cogsci2024"}]
+math = true
+highlight = true
+[header]
+# image = "publications/brockbank2024monster.png"
+caption = ""
++++

content/publication/cao2023semantics.md

@@ -8,10 +8,10 @@
 
 title = "A Semantics for Causing, Enabling, and Preventing Verbs Using Structural Causal Models"
 date = "2023-05-10"
-authors = ["A. Cao","A. Geiger","E. Kreiss","T. Icard","T. Gerstenberg"]
+authors = ["A. Cao\\*","A. Geiger\\*","E. Kreiss\\*","T. Icard","T. Gerstenberg"]
 publication_types = ["3"]
 publication_short = "_Proceedings of the 45th Annual Conference of the Cognitive Science Society_"
-publication = "Cao A., Geiger A., Kreiss E., Icard T., Gerstenberg T. (2023). A Semantics for Causing, Enabling, and Preventing Verbs Using Structural Causal Models. In _Proceedings of the 45th Annual Conference of the Cognitive Science Society_."
+publication = "Cao A.\\*, Geiger A.\\*, Kreiss E.\\*, Icard T., Gerstenberg T. (2023). A Semantics for Causing, Enabling, and Preventing Verbs Using Structural Causal Models. In _Proceedings of the 45th Annual Conference of the Cognitive Science Society_."
 abstract = "When choosing how to describe what happened, we have a number of causal verbs at our disposal. In this paper, we develop a model-theoretic formal semantics for nine causal verbs that span the categories of CAUSE, ENABLE, and PREVENT. We use structural causal models (SCMs) to represent participants' mental construction of a scene when assessing the correctness of causal expressions relative to a presented context. Furthermore, SCMs enable us to model events relating both the physical world as well as agents' mental states. In experimental evaluations, we find that the proposed semantics exhibits a closer alignment with human evaluations in comparison to prior accounts of the verb families."
 image_preview = ""
 selected = false

content/publication/du2024robotic.md

@@ -0,0 +1,33 @@
++++
+# 0 -> 'Forthcoming',
+# 1 -> 'Preprint',
+# 2 -> 'Journal',
+# 3 -> 'Conference Proceedings',
+# 4 -> 'Book chapter',
+# 5 -> 'Thesis'
+
+title = "To Err is Robotic: Rapid Value-Based Trial-and-Error during Deployment"
+date = "2024-06-26"
+authors = ["M. Du","A. Khazatsky","T. Gerstenberg","C. Finn"]
+publication_types = ["2"]
+publication_short = "_arXiv_"
+publication = "Du, M., Khazatsky, A., Gerstenberg, T., Finn, C. (2024). To Err is Robotic: Rapid Value-Based Trial-and-Error during Deployment. _arXiv_."
+abstract = "When faced with a novel scenario, it can be hard to succeed on the first attempt. In these challenging situations, it is important to know how to retry quickly and meaningfully. Retrying behavior can emerge naturally in robots trained on diverse data, but such robot policies will typically only exhibit undirected retrying behavior and may not terminate a suboptimal approach before an unrecoverable mistake. We can improve these robot policies by instilling an explicit ability to try, evaluate, and retry a diverse range of strategies. We introduce Bellman-Guided Retrials, an algorithm that works on top of a base robot policy by monitoring the robot's progress, detecting when a change of plan is needed, and adapting the executed strategy until the robot succeeds. We start with a base policy trained on expert demonstrations of a variety of scenarios. Then, using the same expert demonstrations, we train a value function to estimate task completion. During test time, we use the value function to compare our expected rate of progress to our achieved rate of progress. If our current strategy fails to make progress at a reasonable rate, we recover the robot and sample a new strategy from the base policy while skewing it away from behaviors that have recently failed. We evaluate our method on simulated and real-world environments that contain a diverse suite of scenarios. We find that Bellman-Guided Retrials increases the average absolute success rates of base policies by more than 20% in simulation and 50% in real-world experiments, demonstrating a promising framework for instilling existing trained policies with explicit trial and error capabilities. For evaluation videos and other documentation, go to https://sites.google.com/view/to-err-robotic/home"
+image_preview = ""
+selected = false
+projects = []
+#url_pdf = "papers/du2024robotic.pdf"
+url_preprint = "https://arxiv.org/pdf/2406.15917"
+url_code = ""
+url_dataset = ""
+url_slides = ""
+url_video = ""
+url_poster = ""
+url_source = ""
+url_custom = [{name = "Project website", url = "https://sites.google.com/view/to-err-robotic/home"}]
+math = true
+highlight = true
+[header]
+# image = "publications/du2024robotic.png"
+caption = ""
++++

content/publication/franken2024rails.md

@@ -0,0 +1,33 @@
++++
+# 0 -> 'Forthcoming',
+# 1 -> 'Preprint',
+# 2 -> 'Journal',
+# 3 -> 'Conference Proceedings',
+# 4 -> 'Book chapter',
+# 5 -> 'Thesis'
+
+title = "Procedural dilemma generation for evaluating moral reasoning in humans and language models"
+date = "2024-04-17"
+authors = ["J. Fränken","K. Gandhi","T. Qiu","A. Khawaja","N. D. Goodman","T. Gerstenberg"]
+publication_types = ["3"]
+publication_short = "_Proceedings of the 46th Annual Conference of the Cognitive Science Society_"
+publication = 'Fränken J., Gandhi K., Qiu T., Khawaja A., Goodman N. D., Gerstenberg T. (2024). Procedural dilemma generation for evaluating moral reasoning in humans and language models. In _Proceedings of the 46th Annual Conference of the Cognitive Science Society_.'
+abstract = "As AI systems like language models are increasingly integrated into decision-making processes affecting people's lives, it's critical to ensure that these systems have sound moral reasoning. To test whether they do, we need to develop systematic evaluations. We provide a framework that uses a language model to translate causal graphs that capture key aspects of moral dilemmas into prompt templates. With this framework, we procedurally generated a large and diverse set of moral dilemmas---the OffTheRails benchmark---consisting of 50 scenarios and 400 unique test items. We collected moral permissibility and intention judgments from human participants for a subset of our items and compared these judgments to those from two language models (GPT-4 and Claude-2) across eight conditions. We find that moral dilemmas in which the harm is a necessary means (as compared to a side effect) resulted in lower permissibility and higher intention ratings for both participants and language models. The same pattern was observed for evitable versus inevitable harmful outcomes. However, there was no clear effect of whether the harm resulted from an agent's action versus from having omitted to act. We discuss limitations of our prompt generation pipeline and opportunities for improving scenarios to increase the strength of experimental effects."
+image_preview = ""
+selected = false
+projects = []
+url_pdf = "papers/franken2024rails.pdf"
+url_preprint = "https://arxiv.org/abs/2404.10975"
+url_code = ""
+url_dataset = ""
+url_slides = ""
+url_video = ""
+url_poster = ""
+url_source = ""
+url_custom = [{name = "Github", url = "https://github.com/cicl-stanford/moral-evals/tree/main"}]
+math = true
+highlight = true
+[header]
+# image = "publications/franken2024rails.png"
+caption = ""
++++