← Back to all news
arXiv AI By Priyanka Mary Mammen, Emil Joswin, Srujananjali Medicherla

Do Agents Know When They Succeed? Calibrating Agent Confidence from Internal Representations

Read the original on arXiv AI →

The paper examines whether internal representations of agentic systems can better indicate task success than traditional confidence measures. It introduces two methods—Latent Trajectory Dynamics (LTD) and Action Representation Probe (ARP)—that analyze changes in residual-stream representations and action-level representations, respectively. Experiments on Bash, SQL, and Python benchmarks with Qwen and DeepSeek models show these methods outperform conventional surface-level and sequence-based calibration baselines, offering a zero‑overhead reliability monitor without prompt changes or multiple rollouts.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv AI
Aug 18

From Sequence to Structure: Relational Uncertainty Propagation for LLM Agents

The paper introduces RUPA, a trajectory‑level uncertainty quantification framework for large language model agents. RUPA models an agent’s execution as a directed graph of reasoning states, tool interactions, and environment feedback, then propagates uncertainty across this graph to capture long‑range dependencies. Experiments on benchmarks such as τ‑2, Terminal‑Bench‑2, and GAIA show that RUPA outperforms existing methods, enabling earlier failure detection and more reliable agent execution.

By Zhengzhao Ma. Boxi Cao, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun
llmsagentsbenchmarkssafety
More like this →