This work highlights the need to audit feature transferability and report worst-case domain robustness in cognitive speech screening by proposing a fusion method that integrates XLM-R text baseline scores with evidence anchors selected during training.
Zi-Jian Lu, Si-Zhe Liu, Yin Zhang et al.· 0 citations
This work introduces WirelessOptBench, a benchmark for action assurance in wireless operations that turns wireless tasks into execution state decision episodes with controlled telemetry faults and action constraints, and develops WirelessOpsAgent, which grounds candidate actions in current evidence and repairs recovera...
Zi-Jian Lu, Yi-Ping Zuo, Hao Xu et al.· 1 citation
The results demonstrate the feasibility of a stateful contactless sensing-to-action architecture for long-term home health monitoring and integrates sensing, temporal state, reasoning, and action into a unified and auditable loop.
Xu-Wen Zhang, Zi-Jian Lu, Yi-Cheng Lei et al.· 0 citations
Vision-language models often use descriptions of earlier visual states to make decisions about the current scene. When the scene changes, stale language can redirect an otherwise correct visual judgment toward an outdated answer. We study this failure as visual lock-in in a controlled grounding setting where only the v...
VITAL-RAG is introduced, which organizes evidence by canonical code object, keeps one query-relevant companion only when it adds semantics not already represented, and renders selected evidence under per-object and global token budgets.