Agents used over time encounter recurring professional work: each case requires different evidence and judgment, while the underlying workflow can be reused. Benchmarks built from independent tasks cannot reveal whether an agent turns earlier experience into better procedures for later cases. We introduce FinEvo-Bench,...
DeFA, a dependency-guided framework for agent failure attribution, achieves the highest responsible-agent and exact step accuracy with all evaluated backbones, and the highest failure-mode accuracy among taxonomy-aligned methods on Pro.
Bo Deng, Xin-Lei Zheng, Yi-Xun Wei et al.· 2 citations· ⚡1
Vision-language-action (VLA) models perform well on shorter-horizon manipulation tasks but still struggle with long-horizon tasks that require multiple dependent manipulations from a single command. Online reinforcement learning (RL) can improve these policies through environment interaction, yet many existing methods...
Zi-Yi Yin, Sang-Min Woo, Kang Zhou et al.· 0 citations
FinPerMA is introduced, an event-grounded benchmark that evaluates personalized memory against frozen longitudinal investor trajectories and shows that summary-based memory often preserves factual details while losing the preference signals needed for personalization; simple retrieval can therefore outperform purpose-b...
Ben Wang, Kang Zhou, Li-Fan Guo et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.