Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade
Large language model (LLM) agents solving multi-step tasks frequently commit to trajectories that are doomed to fail, yet continue to consume substantial inference compute before the failure becomes observable. We show that failure is predictable early from the agent's internal representations: lightweight per-round probes on hidden activations anticipate eventual episode failure as early as the first interaction round, where scorers reading only the agent's observable behavior are barely better than chance. We turn this signal into a practical abort cascade: one distribution-free calibrated g
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
- Linked via arxiv authorKai Ruan →
Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade
- Linked via arxiv authorZihe Huang →
Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade
- Linked via arxiv authorZiqi Zhou →
Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade
- Linked via arxiv authorQianshan Wei →
Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade
- Linked via arxiv authorZixuan Wang →
Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade
- Linked via arxiv authorWenhao Sun →
Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade