paper.dinq.meAI 与前沿科学顶会 · 学术新星雷达
登录 DINQ
返回新星榜
LT

Leila Tavakoli

SIGIR 2026×2检索与知识评测与基准可解释性数据与预训练
新星分新星
70
一作贡献40
核心/参与16
方向广度8
合作网络3
1 一作 · 2 篇 · SIGIR 2026

获取完整研究档案

首次获取该候选人的完整档案及 StartLux 评估共 20 积分;每位用户单独解锁,之后查看免费。主动刷新 20 积分。

联系方式

实时查找电子邮箱,费用不超过20积分;未找到邮箱不扣费。

2026 顶会论文

2
一作SIGIR 2026Oral· Information Retrieval (Perspective)

Deletion Isn't Enough: Auditing RAG for Selective Forgetting

Leila Tavakoli · Mark Sanderson

摘要

\begin{abstract}For information access systems, it is not enough for outputs or answers to be relevant or correct; they must also be \emph{permitted}. This paper highlights a research gap: non-disclosure obligations often concern \emph{propositions} that must not be stated, while deployed Retriev...

检索与知识可解释性评测与基准OpenReview
作者位次 #2/3SIGIR 2026Oral· Information Retrieval (Industry)

Policy-Guided RAG: Enforcing Verbatim and Controlled Synthesis

Mina Naghash Asadi · Leila Tavakoli · Mustafa Bilgrami

摘要

Retrieval-Augmented Generation (RAG) typically assumes retrieved text can be freely paraphrased, but this fails in regulated domains where some passages must be quoted verbatim. We propose a policy-guided transformation pattern for RAG, where retrieved segments carry sensitivity metadata and a li...

检索与知识评测与基准数据与预训练OpenReview

常见合作者