Research experiments, one folder each: experiments/<short-name>/. Each one takes a section of the running project
(notebooks/project/retweet_thread.py, one synthetic X world carried through the course) and asks the open question
behind it.
conformal-shiftDo conformal guarantees hold when bots change tactics?Split conformal prediction promises that a prediction set contains the true label with probability at least 1 − α, for any model, as long as calibration and test data are exchangeable. A moderation team that publishes "95% coverage" makes that promise about nedesign-transferAre the best GNN design choices on Cora also the best for bot detection?L19's answer to "which GNN should I use?" is: find a known task whose design ranking resembles yours and reuse its best design. Benchmarks are mostly homophilous citation graphs. Is bot detection on a retweet graph such a task, or does the benchmark's ranking honest-splitsHow much does a random split overestimate a bot detector?Graph benchmarks split nodes at random. In a coordination problem that means members of the same ring land on both sides of the split, and anything the model learns about one member (its targets, its partners, its label if the model uses neighbours' labels) trmotif-signaturesWhich small patterns give rings away, and can a few known accounts find the rest?The project notebook found reciprocal dyads and mutual triangles over-represented, mostly inside rings. But rings can be built without reciprocity (co-amplification), and real coordination research uses a different family of patterns: co-retweet and synchronisrecsys-amplificationDoes the recommender amplify the rings, and by how much?A coordinated ring does not need to fool people directly. It is enough to fool the ranking: if the ring's retweets make its target accounts look popular, the recommender shows those accounts to real users, who then retweet them for real. How much extra exposurrege-vs-wlRegular equivalence vs. Weisfeiler–Lehman for coordinated accountsSociology's regular equivalence (White & Reitz 1983; CATREGE, Borgatti & Everett 1993) and graph ML's colour refinement (1-WL, the ceiling of message-passing GNNs) are the same iterative procedure with one difference: regular equivalence refines by the set of rings-in-communitiesDo coordinated rings look like tiny communities, or hide inside real ones?Community detection is the oldest structural tool of social network analysis, and moderation teams still use it first: "find the dense cluster, look inside". A coordinated ring is dense by construction, but it also points almost all of its retweets at a few po