This work argues that finding people from a natural-language description is fundamentally a requirements engineering task: such a request is an under-determined requirement with implicit constraints, many valid answers, and no acceptance criterion, so useful answers require eliciting, validating, and verifying the requ...
Yuan-Peng He, Fang-Chen Li, Xiang-Yu Ru et al.· 0 citations
A lightweight tool-augmented framework based on dual-teacher distillation and sparse mixture-of-experts (MoE) modeling is proposed to support evidence localization, anomaly prioritization, and expert review in local deployment settings.
This paper examines a complementary design and proposes DenialRAG, a single-document poisoning attack that explicitly names the correct answer, denies it, and presents an attacker-controlled explanation for favoring the wrong answer inside the same retrieved passage.
Abay Zhurekbay, Tao Liu, Fan Li· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
