LLM Relational Event Models

---

Documentation for package ‘llrem’ version 0.1.1

• DESCRIPTION file.
• Package NEWS.

Help Pages

attach_covariates_df	Attach exogenous covariates to a pre-built risk-set data frame
compare_rem_models	Forest plot comparing fitted REM coefficients
compute_hazard_surface	Compute the full hazard surface over all directed dyads
compute_seq_stat_means	Compute sufficient-statistic means for a sequence without fitting a REM
cov_dyad_static	Create a static dyad-level covariate
cov_dyad_temporal	Create a temporal dyad-level covariate
cov_node_static	Create a static node-level covariate
cov_node_temporal	Create a temporal node-level covariate
edgelist	Militarized Interstate Dispute conflict edgelist
fit_rem	Fit a Relational Event Model to an edgelist
fit_rem_sep	Fit a Separable Relational Event Model
llm_call	Call an LLM provider
llm_provider_anthropic	Create an Anthropic Claude LLM provider
llm_provider_bedrock	Create an AWS Bedrock LLM provider
llm_provider_gemini	Create a Google Gemini LLM provider
llm_provider_grok	Create an xAI Grok LLM provider
llm_provider_mock	Create a mock LLM provider for testing
llm_provider_ollama	Create a local or remote Ollama LLM provider
llm_provider_openai	Create an OpenAI LLM provider
llm_provider_openai_compat	Create a generic OpenAI-compatible LLM provider
make_behavior_queries	Generate plain-language behavioral guidelines from empirical sufficient stats
make_query_predict	Query factory: LLM predicts the next event
make_query_roleplay_random	Query factory: LLM roleplays as a randomly assigned node
make_query_roleplay_sender	Query factory: LLM roleplays as the empirical sender
make_stat_means	Compute per-term mean sufficient statistics for realized events
make_suffstats	Extract sufficient-statistic means from a REM risk-set data frame
mock_strategy_highest_indegree	Mock strategy: pick the valid target with the highest in-degree
mock_strategy_max_id	Mock strategy: always pick the largest valid target
mock_strategy_min_id	Mock strategy: always pick the smallest valid target
n_nodes	Number of nodes in the MID edgelist
prepare_edgelist	Normalise a raw edgelist to the standard three-column format
rem_cfg	Create a prompt configuration object
rem_covariates	Collect covariate objects for use in fit_rem
run_llm_rem	Generate an LLM event sequence and fit a REM
run_multiagent_rem	Run a multi-agent REM-LLM conflict simulation