Sparse-reward variant
rl-long-horizon-sparse.jsonSame problem, binary 0/1 success, no intermediate steps.
No sample yet. Add data/trajectories/rl-novel-llm-capabilities/rl-long-horizon-sparse.json.
Same problem, binary 0/1 success, no intermediate steps.
No sample yet. Add data/trajectories/rl-novel-llm-capabilities/rl-long-horizon-sparse.json.