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

Caixia YUAN

ICLR 2026×2ACL 2026可解释性多模态理论与学习动力学检索与知识长上下文与架构
新星分贡献者
43
一作贡献0
核心/参与15
方向广度13
合作网络3
0 一作 · 3 篇 · ICLR 2026 / ACL 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

3
作者位次 #4/5ICLR 2026Poster· Theory / Interpretability and Visualization

An Information-Theoretic Parameter-Free Bayesian Framework for Probing Labeled Dependency Trees from Attention Score

Hongxu Liu · Jing Ma · Xiaojie Wang · Caixia YUAN · Fangxiang Feng

可解释性多模态理论与学习动力学OpenReview
作者位次 #4/5ICLR 2026Poster· Theory / Interpretability and Visualization

Reading Images Like Texts: Sequential Image Understanding in Vision-Language Models

Yueyan Li · Chenggong Zhao · Zeyuan Zang · Caixia YUAN · Xiaojie Wang

多模态可解释性理论与学习动力学OpenReview
作者位次 #4/5ACL 2026Findings· Findings of ACL

E vo M em KG : An Evolvable Memory Agent for Multi-hop Knowledge Graph Reasoning

Shiyu Tian · Shuyue Xing · Zhuoxin Han · Caixia YUAN · Xiaojie Wang

摘要

Integrating knowledge graphs (KGs) with large language models (LLMs) enhances factual accuracy and interpretability in question answering. However, existing agent-based methods rely on static memory mechanisms that fail to address the combinatorial explosion of search spaces in multi-hop reasonin...

检索与知识可解释性长上下文与架构OpenReview 项目页

常见合作者