A unit-by-unit reading of one explicitly versioned Stanford CS224V offering: understanding, dialogue management, generation, evaluation, and deployment for conversational assistants.
CS224V only became Agentic AI in the 2026–2027 catalog, and the rename changed nothing underneath: the course still translates natural language into formal semantics and constrains agents with SMT solvers and knowledge graphs instead of wiring frameworks together. Seven of the eleven mandatory readings come out of the instructor's own lab. Every slide deck is public, and the course site says outright that they are deliberately incomplete.
Fall 2025 opens with computational thinking: reliability comes from decomposing retrieval, formal representation, verification, and generation into testable algorithms, not from one heroic prompt.
STORM uses perspective-guided questions, simulated interviews, and outlines to broaden research; Co-STORM keeps a person in the loop so discovering unknown questions and co-editing become part of the system.
Genie Worksheets declare task capability as a form-like specification. A contextual semantic parser updates formal dialogue state while the runtime controls queries, actions, and responses.
CS224V splits task-agent evaluation into state updates and complete interaction: isolate the semantic parser, then test task completion, grounded queries, and valid actions with real users.
The [WikiChat paper](https://aclanthology.org/2023.findings-emnlp.157/) expands RAG into query formulation, retrieval, filtering, generation, claim extraction, renewed retrieval and verification, and removal of unsupported content—and evaluates retrieval separately from factuality.
SUQL adds answer and summary functions over text to SQL. A semantic parser emits one hybrid query, while an optimizing compiler applies predicate pushdown, top-k pruning, and lazy evaluation.
SLIDERS induces a question-specific schema, applies semantic chunking and contextualized extraction, reconciles duplicate rows, and answers with SUQL instead of feeding every long document directly to one model.
Automated qualitative coding defines event types and arguments in a codebook, then separates document classification, structured extraction, and entity linking. Constrained JSON fixes form, not expert judgment.
SPINACH does not guess complete SPARQL in one shot. It searches entities and properties, inspects Wikidata entries and examples, executes small queries, and composes a final query under explicit action and stopping rules.
The lecture parses patient records and trial criteria into SMT, retrieves candidates through a weaker propositional projection, and runs a solver on the reduced set. Reasoning is inspectable, but NL-to-SMT remains the main error boundary.
CHURRO represents full-page text, layout, and metadata in HDML, unifies multilingual historical data for a page-level VLM, and connects extraction to HistoryGenie for searchable, conversational archives.
ReactGenie annotates React components to expose data, actions, and views, parses composite voice commands into a DSL, and renders native graphical output against shared UI context.
The final lecture is not a complete LLM-training tutorial. It studies data efficiency under fixed data and abundant compute, revisiting epochs, batches, ensembles, self-training, and conditions for synthetic continued pretraining.