Schedule Spring Quarter 2023/24

Week 1,  Apr 17th

Topic: LLM agents

Papers:

• Wu, Qingyun, et al. "Autogen: Enabling next-gen llm applications via multi-agent conversation framework." arXiv preprint arXiv:2308.08155 (2023). https://arxiv.org/abs/2308.08155 

Discussion lead: Chi Wang (guest)

Week 2, Apr 24th

Topic: Prediction-powered inference and LLM evaluation

Papers:

• Angelopoulos, A.N., Bates, S., Fannjiang, C., Jordan, M.I. and Zrnic, T., 2023. Prediction-powered inference. Science, 382(6671), pp.669-674. https://www.science.org/doi/full/10.1126/science.adi6000 

• Zrnic, T. and Candès, E.J., 2024. Active Statistical Inference. arXiv preprint arXiv:2403.03208.  https://arxiv.org/abs/2403.03208 

• Boyeau, P., Angelopoulos, A.N., Yosef, N., Malik, J. and Jordan, M.I., 2024. AutoEval Done Right: Using Synthetic Data for Model Evaluation. arXiv preprint arXiv:2403.07008. https://arxiv.org/abs/2403.07008

Discussion lead: Tijana Zrnić (guest)

Week 3, May 1st

Topic: Automated data testing in Australian politics and Canadian journalism

Papers:

• Digitization of the Australian Parliamentary Debates, 1998–2022 https://www.nature.com/articles/s41597-023-02464-w 

• Digitization of the Canadian Parliamentary Debates https://www.jstor.org/stable/26858546 

• Implementing Automated Data Validation for Canadian Political Datasets https://arxiv.org/abs/2309.12886 

• Evaluating the Decency and Consistency of Data Validation Tests Generated by LLMs https://arxiv.org/abs/2310.01402 

Discussion lead: Lindsay Katz (guest)

Week 4, May 8th

Topic: Concept-aware LLMs

Papers: 

• Shani, Chen, Jilles Vreeken, and Dafna Shahaf. "Towards Concept-Aware Large Language Models." https://aclanthology.org/2023.findings-emnlp.877/

• Optional:

• Jiang, Yang, et al. "Expert feature-engineering vs. deep neural networks: which is better for sensor-free affect detection?." Artificial Intelligence in Education: 19th International Conference, AIED 2018, London, UK, June 27–30, 2018, Proceedings, Part I 19. Springer International Publishing, 2018. https://learninganalytics.upenn.edu/ryanbaker/jiang-aied2018.pdf 

• Rytting, Christopher, and David Wingate. "Leveraging the inductive bias of large language models for abstract textual reasoning." Advances in Neural Information Processing Systems 34 (2021): 17111-17122. https://arxiv.org/pdf/2110.02370 

• Si, Chenglei, et al. "Measuring inductive biases of in-context learning with underspecified demonstrations." arXiv preprint arXiv:2305.13299 (2023). https://aclanthology.org/2023.acl-long.632.pdf

• Papadimitriou, Isabel, and Dan Jurafsky. "Injecting structural hints: Using language models to study inductive biases in language learning." Findings of the Association for Computational Linguistics: EMNLP 2023. 2023. https://aclanthology.org/2023.findings-emnlp.563.pdf

• Cazenavette, George, and Simon Lucey. "On the bias against inductive biases." arXiv preprint arXiv:2105.14077 (2021). https://arxiv.org/pdf/2105.14077 

Discussion lead: Chen Shani

Week 5, May 15th

Topic: European Union Framework for Governing Generative AI Models

Papers: 

• An Overview of the European Union Framework Governing Generative AI Models and Systems. Florence G'sell. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4762804 

Discussion lead: Florence G'Sell (guest)

Week 6, May 22nd

Topic:  LM-mediated information and persuasion

Papers: 

• https://mit-genai.pubpub.org/pub/hu71se89/release/1 

• https://www.anthropic.com/research/measuring-model-persuasiveness 

• https://academic.oup.com/pnasnexus/article/3/2/pgae034/7610937?login=true 

• https://assets-eu.researchsquare.com/files/rs-3238396/v1_covered_61d6cb87-346c-47a0-a741-01c5a191e518.pdf?c=1694141729 

• https://dl.acm.org/doi/pdf/10.1145/3613904.3642076 

Discussion lead: Isabel Gallegos

Week 7, May 29th

Topic:  AI and Labor

Papers: 

• Gen-AI: Artificial Intelligence and the Future of Work https://www.imf.org/en/Publications/Staff-Discussion-Notes/Issues/2024/01/14/Gen-AI-Artificial-Intelligence-and-the-Future-of-Work-542379 

• The impact of artificial intelligence on growth and employment https://cepr.org/voxeu/columns/impact-artificial-intelligence-growth-and-employment 

• Forecasting extreme labor displacement: A survey of AI practitioners https://www.sciencedirect.com/science/article/pii/S0040162520311495

Optional:

• GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models https://arxiv.org/abs/2303.10130 

• Which U.S. Workers Are More Exposed to AI on Their Jobs? https://www.pewresearch.org/social-trends/2023/07/26/which-u-s-workers-are-more-exposed-to-ai-on-their-jobs/ 

Discussion lead: Yifan Mai

Schedule Winter Quarter 2023/24

Week 1,  Jan 24th

Topic: Probing for making consistent measurements with LLMs

Papers:

• Unsupervised Contrast-Consistent Ranking with Language Models (Stoehr et al. 2023)

• Characterizing Mechanisms for Factual Recall in Language Models (Yu et al., 2023)

Optional:

• Measurement in the Age of LLMs: An Application to Ideological Scaling (O’Hagan and Schein, 2023)

• Discovering Latent Knowledge in Language Models Without Supervision (Burns et al., 2023)

• Linear Representations of Sentiment in Large Language Models (Tigges et al, 2023)

Discussion lead: Niklas Stoehr (guest)

Week 2,  Jan 31st

Topic: Persuasion in the era of LLMs

Papers:

• How Johnny Can Persuade LLMs to Jailbreak Them: Rethinking Persuasion to Challenge AI Safety by Humanizing LLMs

• The Earth is Flat because...: Investigating LLMs' Belief towards Misinformation via Persuasive Conversation

Discussion lead: Weiyan Shi

Week 3,  Feb 14th

Topic: NLP & Policing

Papers:

• Rho, Eugenia H., et al. "Escalated police stops of Black men are linguistically and psychologically distinct in their earliest moments." Proceedings of the National Academy of Sciences 120.23 (2023): e2216162120. https://www.pnas.org/doi/full/10.1073/pnas.2216162120 

Optional:

• Pierson, E., Simoiu, C., Overgoor, J., Corbett-Davies, S., Jenson, D., Shoemaker, A., ... & Goel, S. (2020). A large-scale analysis of racial disparities in police stops across the United States. Nature human behaviour, 4(7), 736-745. https://www.nature.com/articles/s41562-020-0858-1 

• Weisburd, D., Telep, C. W., Vovak, H., Zastrow, T., Braga, A. A., & Turchan, B. (2022). Reforming the police through procedural justice training: A multicity randomized trial at crime hot spots. Proceedings of the National Academy of Sciences, 119(14), e2118780119. https://www.pnas.org/doi/full/10.1073/pnas.2118780119 

• Prowse, G., Weaver, V. M., & Meares, T. L. (2020). The state from below: Distorted responsiveness in policed communities. Urban Affairs Review, 56(5), 1423-1471. https://journals.sagepub.com/doi/full/10.1177/1078087419844831 

Discussion lead: Maggie Harrington

Week 4,  Feb 21st

Topic: Values in LLMs

Papers:

• Value Kaleidoscope: Engaging AI with Pluralistic Human Values, Rights, and Duties https://arxiv.org/abs/2309.00779 

• A Roadmap to Pluralistic Alignment https://arxiv.org/abs/2402.05070 

• LLMs grasp morality in concept https://arxiv.org/abs/2311.02294 

• Optional:

  • Can Machines Learn Morality? The Delphi Experiment. https://arxiv.org/abs/2110.07574

  • Aligning AI With Shared Human Values https://arxiv.org/abs/2008.02275  

Discussion lead: Jared Moore

Week 5,  Feb 28th

Topic: Reasoning about networks with LLMs

Papers:

• Papachristou, Marios, and Yuan Yuan. "Network Formation and Dynamics Among Multi-LLMs." arXiv preprint arXiv:2402.10659 (2024). https://arxiv.org/pdf/2402.10659.pdf 

• Fatemi, Bahare, Jonathan Halcrow, and Bryan Perozzi. "Talk like a graph: Encoding graphs for large language models." arXiv preprint arXiv:2310.04560 (2023). https://arxiv.org/pdf/2310.04560.pdf 

Discussion lead: Alicja Chaszczewicz

Week 6,  Mar 6th

Topic: Beyond Memorization: Stronger Language Models and Their Privacy Implication

Papers:

• Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theory https://arxiv.org/abs/2310.17884 

• What Does it Mean for a Language Model to Preserve Privacy? https://arxiv.org/abs/2202.05520 

• Optional:

  • R-Judge: Benchmarking Safety Risk Awareness for LLM Agents https://arxiv.org/abs/2401.10019 

Discussion lead: Yijia Shao

Schedule Fall Quarter 2023/24

Week 1, Oct 16th

Topic: Language, Identity, and Language Model Interaction

Papers:

• Schlesinger, Ari, W. Keith Edwards, and Rebecca E. Grinter. "Intersectional HCI: Engaging identity through gender, race, and class." Proceedings of the 2017 CHI conference on human factors in computing systems. 2017. https://dl.acm.org/doi/10.1145/3025453.3025766

• Zamfirescu-Pereira, J. D., et al. "Why Johnny can’t prompt: how non-AI experts try (and fail) to design LLM prompts." Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems. 2023. https://dl.acm.org/doi/abs/10.1145/3544548.3581388 

• Optional:

  • Zheng, Lianmin, et al. "LMSYS-Chat-1M: A Large-Scale Real-World LLM Conversation Dataset." arXiv preprint arXiv:2309.11998 (2023). https://arxiv.org/pdf/2309.11998v3.pdf 

  • Volkova, Svitlana, et al. "Inferring latent user properties from texts published in social media." Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 29. No. 1. 2015. https://ojs.aaai.org/index.php/AAAI/article/view/9271 

Discussion lead: Will Held

Week 2, Oct 23rd

Topic: From Question Answering to Question Asking

Papers:

• Large Language Models for Automated Open-domain Scientific Hypotheses Discovery (arxiv preprint, https://arxiv.org/abs/2309.02726)

• Socially situated artificial intelligence enables learning from human interaction (PNAS Vol. 119 | No. 39, https://www.pnas.org/doi/full/10.1073/pnas.2115730119 )

• Don’t Just Tell Me, Ask Me: AI Systems that Intelligently Frame Explanations as Questions Improve Human Logical Discernment Accuracy over Causal AI explanations (CHI 2023, https://dl.acm.org/doi/10.1145/3544548.3580672)

Discussion lead: Yijia Shao

Week 3, Oct 30th

Topic: Anthropomorphism

Papers:

• See in the slack channel.

Optional

• Talking About Large Language Models https://arxiv.org/abs/2212.03551

• Computational analysis of 140 years of US political speeches reveals more positive but increasingly polarized framing of immigration https://www.pnas.org/doi/epdf/10.1073/pnas.2120510119

Discussion lead: Myra Cheng

Week 4, Nov 6th

Topic: Guiding Design of LLM-powered chatbots with Expert Feedback

Papers:

• Petridis, Savvas, et al. "ConstitutionMaker: Interactively Critiquing Large Language Models by Converting Feedback into Principles." https://arxiv.org/abs/2310.15428 

• Chen, Siyuan, et al. "LLM-empowered Chatbots for Psychiatrist and Patient Simulation: Application and Evaluation." https://arxiv.org/abs/2305.13614 

Discussion lead: Ryan Louie

Week 5, Nov 13th

Topic: Representation of global culture and values in LLMs

Papers:

• Towards Measuring the Representation of Subjective Global Opinions in Language Models https://arxiv.org/pdf/2306.16388.pdf  

• Having Beer after Prayer? Measuring Cultural Bias in Large Language Models https://arxiv.org/pdf/2305.14456.pdf  

• (only Section 5) BHASA: A Holistic Southeast Asian Linguistic and Cultural Evaluation Suite for Large Language Models https://arxiv.org/vc/arxiv/papers/2309/2309.06085v1.pdf  

Discussion lead: Yifan Mai

Week 6, Nov 27th

Topic: Conversational grounding with LLMs

Papers:

• Grounding or Guesswork? Large Language Models are Presumptive Grounders https://arxiv.org/abs/2311.09144 

• Principles of Mixed-Initiative Interaction https://dl.acm.org/doi/pdf/10.1145/302979.303030 

• Zero-Shot Goal-Directed Dialogue via RL on Imagined Conversations https://arxiv.org/abs/2311.05584 

• Optional:

  • Co-constructing intersubjectivity with artificial conversational agents: People are more likely to initiate repairs of misunderstandings with agents represented as human https://www.sciencedirect.com/science/article/pii/S0747563215303101

  • Generative Theories of Interaction https://hal.science/hal-03434142/file/GenTheory%20authorversion.pdf 

Discussion lead: Omar Shaikh

Week 7, Dec 4th

Topic: Eliciting human preferences with Language models

Papers:

• Eliciting Human Preferences with Language Models. Belinda Z. Li, Alex Tamkin, Noah Goodman, Jacob Andreas. https://arxiv.org/abs/2310.11589 

Discussion lead: Alex Tamkin

Schedule Spring Quarter 2022/23

Week 1, Apr 6th

Topic: Overconfidence and uncertainty

Papers:

• On Hedging in Physician-Physician Discourse; Ellen F. Prince (Linguistics), Joel Frader (Pediatrics), and Charles Bosk (Sociology)." (1980). http://www.cs.columbia.edu/~prokofieva/CandidacyPapers/Prince_Hedging.pdf 

• Zhou, Kaitlyn, Dan Jurafsky, and Tatsunori Hashimoto. "Navigating the Grey Area: Expressions of Overconfidence and Uncertainty in Language Models." arXiv preprint arXiv:2302.13439 (2023). https://arxiv.org/abs/2302.13439 

• Mielke, Sabrina J., et al. "Reducing conversational agents’ overconfidence through linguistic calibration." Transactions of the Association for Computational Linguistics 10 (2022): 857-872. https://arxiv.org/abs/2012.14983 

Discussion lead: Kaitlyn Zhou

Week 2, Apr 13th

Topic: Large language models as simulated humans

Papers:

• Generative Agents: Interactive Simulacra of Human Behavior https://arxiv.org/abs/2304.03442 

• Whose Opinions Do Language Models Reflect? https://arxiv.org/abs/2303.17548 

• (Optional) Large Language Models as Simulated Economic Agents: What Can We Learn from Homo Silicus? https://john-joseph-horton.com/papers/llm_ask.pdf 

Discussion lead: Joon Park

Week 3, Apr 20th

Topic: Culture and pragmatics

• Thomas, Jenny. "Cross-cultural pragmatic failure." Applied linguistics 4.2 (1983): 91-112. 

Discussion lead: Jing Huang

Week 4, Apr 27th

Topic: Reflection and self-correction

• The Capacity for Moral Self-Correction in Large Language Models. https://arxiv.org/abs/2302.07459

• Reflexion: an autonomous agent with dynamic memory and self-reflection. https://arxiv.org/abs/2303.11366

Discussion lead: Tiziano Piccardi

Week 5, May 4th

Topic: Polarization and mitigations

• Engaging Politically Diverse Audiences on Social Media

• Negativity drives online news consumption

• Artificial Intelligence Can Persuade Humans on Political Issues

Discussion lead: Martin Saveski

Week 6, May 11th

Topic: Causality and common sense

• Causal Reasoning and Large Language Models: Opening a New Frontier for Causality. Emre Kıcıman, Robert Ness, Amit Sharma, Chenhao Tan. https://arxiv.org/abs/2305.00050 

• Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMs. Maarten Sap, Ronan Le Bras, Daniel Fried, and Yejin Choi https://arxiv.org/pdf/2210.13312 

Discussion lead: Kristina Gligoric

Week 7, May 18th

Topic: Measuring stereotypes in LLMs and in simulation scenarios

• Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Models. Cheng et al. (pdf in the Slack channel)

Discussion lead: Myra Cheng

Week 8, May 25th

Topic: Large language models and linguistic theory

• What’s the deal with Chomsky? And other commonly asked questions about linguistics

• Chomsky’s The False Promise of ChatGPT

• Modern language models refute Chomsky’s approach to language

• Why large language models are poor theories of human linguistic cognition. A reply to Piantadosi (2023).

Optional

• Modern Language Models Refute Nothing (response on twitter)

• The Impact of Large Language Models on Linguistic Theory and Generative Grammar: A Critical Analysis

Discussion lead: Omar Shaikh

Week 9, Jun 1st

Topic: AI and persuasion

• Generative Language Models and Automated Influence Operations: Emerging Threats and Potential Mitigations. Goldsteain et al. https://arxiv.org/pdf/2301.04246.pdf 

• Quantifying the potential persuasive returns to political microtargeting. Tappin et al. https://psyarxiv.com/dhg6k/ 

• Working with AI to persuade: Examining a large language model’s ability to generate pro-vaccination messages. Karinshak et al. https://dl.acm.org/doi/abs/10.1145/3579592 

• Exploiting Programmatic Behavior of LLMs: Dual-Use Through Standard Security Attacks. Kang et al. https://arxiv.org/pdf/2302.05733.pdf

  Discussion lead: Rylan Schaeffer

Week 10, Jun 8th

Brainstorming session

Schedule Winter Quarter 2022/23

Week 1, Jan 19th

Discussing the logistics

Week 2, Jan 26th

Topic:  Causality

Papers: 

• Feder, Amir, et al. "Causal inference in natural language processing: Estimation, prediction, interpretation and beyond." Transactions of the Association for Computational Linguistics 10 (2022): 1138-1158. https://arxiv.org/pdf/2109.00725.pdf 

• Keith, Katherine A., Douglas Rice, and Brendan O'Connor. "Text as Causal Mediators: Research Design for Causal Estimates of Differential Treatment of Social Groups via Language Aspects." arXiv preprint arXiv:2109.07542 (2021). https://arxiv.org/pdf/2109.07542.pdf 

• Keidar, Daphna, et al. "Slangvolution: A Causal Analysis of Semantic Change and Frequency Dynamics in Slang." arXiv preprint arXiv:2203.04651 (2022).https://arxiv.org/pdf/2203.04651.pdf 

Discussion lead: Kristina Gligoric

Week 3, Feb 2nd

Topic:  Reinforcement learning

Papers: 

• Pyatkin, Valentina, et al. "Reinforced Clarification Question Generation with Defeasibility Rewards for Disambiguating Social and Moral Situations." arXiv preprint arXiv:2212.10409 (2022). https://arxiv.org/abs/2212.10409 

• Ramamurthy, Rajkumar, et al. "Is Reinforcement Learning (Not) for Natural Language Processing?: Benchmarks, Baselines, and Building Blocks for Natural Language Policy Optimization." arXiv preprint arXiv:2210.01241 (2022). https://arxiv.org/abs/2210.01241 

Discussion lead: Caleb Ziems

Week 4, Feb 9th

Topic: Prediction, explanation, and integrating the two in NLP

Papers:

• Shmueli. To Explain or to predict? https://projecteuclid.org/journals/statistical-science/volume-25/issue-3/To-Explain-or-to-Predict/10.1214/10-STS330.full

• Schäfer, Mike S., and Valerie Hase. "Computational methods for the analysis of climate change communication: Towards an integrative and reflexive approach." Wiley Interdisciplinary Reviews: Climate Change (2022): e806. https://wires.onlinelibrary.wiley.com/doi/epdf/10.1002/wcc.806

• Hofman, Jake M., et al. "Integrating explanation and prediction in computational social science." Nature 595.7866 (2021): 181-188. https://www.nature.com/articles/s41586-021-03659-0

• Agrawal, Mayank,  et al. "Scaling up psychology via scientific regret minimization." Proceedings of the National Academy of Sciences 117, no. 16 (2020): 8825-8835. https://www.pnas.org/doi/10.1073/pnas.1915841117

Discussion lead: Myra Cheng

Week 5, Feb 16th

Topic: Common ground and communication 

Papers: 

• Herbert Clark’s “Using Language” (Chapter 4) Chapter 4.pdf

• Contributing to Discourse. https://onlinelibrary.wiley.com/doi/pdfdirect/10.1207/s15516709cog1302_7

• Grounding as a Collaborative Process. https://aclanthology.org/2021.eacl-main.41.pdf

• Optional reading:

  • • Communities, Commonalities, and Communication

    • https://www.sol.lu.se/media/utbildning/dokument/kurser/LINB30/20082/Clark.pdf

    • Shared common ground influences information density in microblog texts

    • https://aclanthology.org/N15-1182.pdf

Discussion lead: Omar Shaikh

Week 6, Feb 23rd

Topic: Human-centered data and language models: Privacy, data as labor, and licensing

Papers:

• Contractor, Danish, et al. "Behavioral use licensing for responsible AI." 2022 ACM Conference on Fairness, Accountability, and Transparency. 2022. https://dl.acm.org/doi/abs/10.1145/3531146.3533143 

• The Paradox of Reuse, Language Models Edition. https://nmvg.mataroa.blog/blog/the-paradox-of-reuse-language-models-edition/ 

• ChatGPT Stole Your Work. So What Are You Going to Do? https://www.wired.com/story/chatgpt-generative-artificial-intelligence-regulation/ 

Guest: Nick Vincent

Week 7, Mar 2nd

Topic: Agreeability among humans

Papers: 

• Bakker, Michiel A., et al. "Fine-tuning language models to find agreement among humans with diverse preferences." https://www.deepmind.com/publications/fine-tuning-language-models-to-find-agreement-among-humans-with-diverse-preferences 

• Argyle, Lisa P., et al. "AI Chat Assistants can Improve Conversations about Divisive Topics." https://arxiv.org/pdf/2302.07268.pdf 

Discussion lead: Nicole Meister

Week 8, Mar 9th

Topic: Simulating humans with LLMs: Challenges and opportunities for social science research

Papers: 

• Argyle, Lisa P., et al. "Out of one, many: Using language models to simulate human samples." Political Analysis (2022): 1-15. https://arxiv.org/pdf/2209.06899.pdf

• Aher, Gati, Rosa I. Arriaga, and Adam Tauman Kalai. "Using Large Language Models to Simulate Multiple Humans." https://arxiv.org/pdf/2208.10264.pdf 

• I strongly feel that this is an insult to life itself, Kevin Munger. https://kevinmunger.substack.com/p/i-strongly-feel-that-this-is-an-insult 

Optional: Kosinski, Michal. "Theory of mind may have spontaneously emerged in large language models." arXiv preprint arXiv:2302.02083 (2023). https://arxiv.org/pdf/2302.02083.pdf 

Discussion lead: Tiziano Piccardi 

Week 9, Mar 16th

Topic: Race and racism

Papers:

• • Voigt, Rob, et al. "Language from police body camera footage shows racial disparities in officer respect." Proceedings of the National Academy of Sciences 114.25 (2017): 6521-6526. https://www.pnas.org/doi/10.1073/pnas.1702413114 

  • Prabhakaran, Vinodkumar, et al. "Detecting institutional dialog acts in police traffic stops." Transactions of the Association for Computational Linguistics 6 (2018): 467-481. https://transacl.org/ojs/index.php/tacl/article/view/1349 

  • Camp, Nicholas P., et al. "The thin blue waveform: Racial disparities in officer prosody undermine institutional trust in the police." Journal of personality and social psychology 121.6 (2021): 1157. https://www.apa.org/pubs/journals/releases/psp-pspa0000270.pdf 

  • Epp, Charles R., Steven Maynard‐Moody, and Donald Haider Markel. "Beyond profiling: The institutional sources of racial disparities in policing." Public Administration Review 77.2 (2017): 168-178. https://onlinelibrary.wiley.com/doi/full/10.1111/puar.12702 

Discussion lead: Anjalie Field