Schedule for CMU 10-718: Spring 2026

Week	Date	Class Type	Topic	Resources	Announcements
1	Tue Jan 13	lecture	Intro and Overview - History, Motivation, Logistics [Slides]	Optional readings H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models, Zhang et al 2023 Efficient streaming LMs with attention sinks, Xiao et al 2024	Feedback to give Give feedback Signup to present Signup
1	Thu Jan 15	lecture	ML Project Deep Dive - Project definition and scoping [Slides]	Optional readings project scoping guide project scoping worksheet	Required readings Artificial Intelligence—The Revolution Hasn’t Happened Yet The bitter lesson
2	Tue Jan 20	lecture	Data Preparation and Visualization - Visualization techniques [Slides]	Optional readings Wordcloud examples, etc Quickdraw NY Times Air Pollution Wind maps FACETS project Quickdraw blog post One QuickDraw project Other QuickDraw data analysis projects Obligatory xkcd slide	Deadlines Signup to present in first 1/3 of course Required readings Debugging Large-Scale Data Science Pipelines using Dagger, Rezig et al, VLDB 2020 Goal Driven Discovery of Distributional Differences via Language Descriptions
2	Thu Jan 22	lecture	Data Preparation and Visualization, Problem Formulation - Deduping and soft-joining data [Slides]		Deadlines Signup for AWS Required readings Dissecting Racial Bias in an Algorithm Used to Manage the Health of Populations Here’s What You Should Know About Launching an AI Startup - WIRED 2025 Here’s What You Should Know ... WIRED (cached version)
3	Tue Jan 27	lecture	Data & Feature Engineering [Slides]		Deadlines Project Assignment 0 due (team and alpha version of proposal) Required readings Pervasive Label Errors in Test Sets Destabilize Machine Learning Benchmarks Can Foundation Models Wrangle Your Data?
3	Thu Jan 29	lecture	Evaluation [Slides]		Required readings Rephrasing the Web: A Recipe for Compute and Data-Efficient Language Modeling, ACL 2024 Amazing Things Come From Having Many Good Models, ICML 2024 Prediction Powered Inference
4	Tue Feb 3	lecture	Hyperparameter tuning [Slides]	Optional readings Greg Kochanski, Google Deep Mind, guest lecture on Vizier, Spring 2025 in 10-605 William's lecture on Hyperparameter Search, Fall 2025 in 10-605 (alternate to Greg's talk) Hyperband - A Novel Bandit Based Approach to Hyperparameter Optimization, JMLR 2018	Required readings Google Vizier: A Service for Black-Box Optimization (KDD 2017) Are Transformers Effective for Time Series Forecasting? (AAAI 2023)
4	Thu Feb 5	lecture	Interpretability and Monitorability [Slides]	Optional readings LMs Don't Always Say What They Think: Unfaithful Explanations in CoT, Turpin et al 2023 Interpretable Traces, Unexpected Outcomes: Bhambri et 2025 Beyond Semantics: Reasonless Reasoning Tokens, Valmeekam et al, 2025 (How) Do reasoning models reason? Kambhampati et al, 2025 Semi-structured LLM Reasoners Can Be Rigorously Audited, Leng et al 2025	Required readings The Mythos of Interpretability, Lipton 2018 Monitoring Monitorability, Guan et al 2025 Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation, Baker et al, 2025
5	Tue Feb 10	lecture	Project Checkins 1		Deadlines Project Assignment 1 due (project checkins)
5	Thu Feb 12	lecture	Project Checkins 2
6	Tue Feb 17	lecture	Interpretability as Communication [Slides]	Optional readings Learning to Decieve with Attention-Based Explanations, 2020 Goodhart’s Law Applies to NLP’s Explanation Benchmarks. 2023 Evaluating Explanations: How much do explanations from the teacher aid students? Pruthi et al 2021. Performative Prediction, PMLR 2020	Required readings Unraveling the Enigma of Reason (Scott Young, blog post) Learning to Scaffold: Optimizing Model Explanations for Teaching, Fernandes et al, 2022 Feedback to give Give feedback
6	Thu Feb 19	lecture	Network Effects and ML [Slides]	Optional readings Information cascades, Ch 16, Easly and Kleinberg, 2010 Salganic, Dodds, Watts, Science 2006 Schelling Spatial Segregation model	Required readings LLM use in scholarly writing, Earp et al, 2025 Artificial Hivemind: The Open-Ended Homogeneity of Language Models, 2025
7	Tue Feb 24	lecture	Project Pitches		Deadlines Project Assignment 2 due (talks for project pitches)
7	Thu Feb 26	lecture	Project Pitches		Deadlines Project Assignment 2 due (writeup, due 2/27)
8	Tue Mar 3	lecture	Spring break 1 - no class
8	Thu Mar 5	lecture	Spring break 2 - no class
9	Tue Mar 10	lecture	Working with Agents [Slides]	Optional readings A simple ReAct agent, Maameri Medium post, 2025 RAG for Knowlege-Intensive NLP Tasks, Lewis et al, NeurIPS 2020 PoT Prompting, Chen et al, TMLR, 2023 Yao et al, ICLR 2023 How Does OpenClaw Work? Symkowski Medium post, 2026 An AI Agent Published a Hit Piece on Me, Shambaugh blog post, 2026 Holistic Agent Leaderboard, GitHub site, 2026 Measuring AI agent autonomy in practice, Anthropic blog post, 2026 Evaluating AGENTS.md, Gloaguen et al, 2026	Required readings “Measuring Agents in Production, 2025" When Do Multi-Agent Systems Fail, 2025
9	Thu Mar 12	lecture	Model Compression - Guest speaker, Steven Kolawole		Required readings Model Compression in Practice, Hohman et al, CHI 2024 What Do Compressed Deep Neural Networks Forget? Hooker et al, 2021
10	Tue Mar 17	lecture	Distribution Shift [Slides]	Optional readings Rethinking Distribution Shifts, Wang et al 2023 Detecting Change in Data Streams, Kifer, Ben-David, Gehrke, VLDB 2004 Frustratingly Easy Domain Adaptation, Hal Daume III, 2007 Adaptation of maximum entropy ..., Chelba and Acero, 2004 fastml.com Blog, 2023	Required readings Unsupervised Domain Adaptation by Backpropagation, Ganin & Lempitsky ICML 2015. The data addition dilemma, Shen et al, 2024
10	Thu Mar 19	lecture	Distribution Shift and Robustness [Slides]	Optional readings Goodfellow et al, Explaining and harnessing adversarial examples, ICLR 2015 Hacking the brain with adversarial images, IEEE Spectrum, 2018 Adversarial Examples that Fool both Computer Vision and Time-Limited Human, Elsayed et al, 2018 Adversarial Examples Are Not Bugs, They Are Features, Ilyas et al, 2019 Eykholt and Evtimov, et al, Robust Physical-World Attacks on Deep Learning Visual Classification, CVPR 2018 Lewis et al, 2025	Required readings Task arithmetic paper, 2023 Interpreting Adversarial Examples in Deep Learning, 2025
11	Tue Mar 24	lecture	Mechanistic Interpretability		Required readings A Pragmatic Vision for Interpretability, Neel Nanda et al The Dead Salmons of AI Interpretability, 2025
11	Thu Mar 26	lecture	Graph Data - Guest Speaker, Karish Grover
12	Tue Mar 31	lecture	Project Presentations - 1		Deadlines Project Assignment 4 due (proposed extensons)
12	Thu Apr 2	lecture	Project Presentations - 2
13	Tue Apr 7	lecture	Causality		Required readings Causal Classification: Treatment Effect Estimation vs. Outcome Prediction, 2022 Learning treatment effects while treating those in need, Bryan Wilder, Pim Welle, 2024
13	Thu Apr 9	lecture	Carnival - no class
14	Tue Apr 14	lecture	Privacy and Federated Learning		Required readings MAGPIE: A benchmark for Multi-AGent contextual PrIvacy Evaluation, Juneja et al, 2025 Federated learning with differential privacy for breast cancer diagnosis, Shukla and Rajkumar, 2025
14	Thu Apr 16	lecture	Bias and Variance in Modern ML		Required readings Base Models Beat Aligned Models at Randomness and Creativity, COLM 2025 Rethinking Cross-lingual Gaps from a Statistical Viewpoint, Piratla et al, 2025
15	Tue Apr 21	lecture	Final Project Presentations - 1		Deadlines Project Assignment 5 due (final writeup and reflections)
15	Thu Apr 23	lecture	Final Project Presentations - 2