亲爱的研友该休息了!由于当前在线用户较少,发布求助请尽量完整地填写文献信息,科研通机器人24小时在线,伴您度过漫漫科研夜!身体可是革命的本钱,早点休息,好梦!

Large language models as tools for Bayesian causal discovery

作者
Bruna Bazaluk
标识
DOI:10.11606/d.45.2026.tde-18032026-165543
摘要

Causal Discovery (CD) is the task of automatically inferring causal structures, typically from obser- vational data. Recently, there has been much interest in utilizing domain knowledge from large language models (LLM) in causal discovery. However, existing LLM-based approaches only output a single directed acyclic graph (DAG) without uncertainty, which can be unreliable. Thus, the focus of the present disserta- tion is exploring Bayesian Structure Learning algorithms which produce distributions of graphs, natural representations of uncertainty and using LLMs to enhance their inference power. The present disser- tation has three main goals. The first is discussing and reviewing the basic algorithms of CD, and also focusing on LLM-based approaches. There is also a short section that explores different methods of defin- ing the LLMs output reliability. Secondly, we intend to investigate the usage of LLMs alongside Bayesian structure learning (BSL) methods for causal discovery. In particular, we propose to harness the domain knowledge from the LLMs as the prior distribution over graphs. Our experiments show that LLM-informed priors can improve the performance of BSL methods. Lastly, this text describes another approach, which focuses on Ancestral Graphs (AGs), where we do not need the sufficiency assumption. Furthermore, our model describes a method that uses the LLM as an expert-in-the-loop, mixing the data information with the outside knowledge during the model processing. Our experiments show that our method is competitive against others that address latent confounding on both synthetic and real-world datasets; and our design for incorporating feedback from a (simulated) human expert or a Large Language Model (LLM) improves inference quality.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
9秒前
科研通AI65的应助被萧雨墨采纳,获得10
12秒前
张阳发布了新的文献求助10
14秒前
19秒前
19秒前
19秒前
复杂的万恶完成签到,获得积分10
21秒前
认真的龙猫完成签到 ,获得积分10
23秒前
28秒前
贱小贱完成签到,获得积分0
28秒前
me6363发布了新的文献求助10
33秒前
139完成签到 ,获得积分10
38秒前
无极微光的应助被行必至采纳,获得20
46秒前
科研通AI6.2的应助被萧雨墨采纳,获得10
53秒前
飘逸的安珊完成签到,获得积分10
58秒前
58秒前
飞快的不尤完成签到,获得积分10
59秒前
Jazz33333333完成签到,获得积分10
59秒前
1874发布了新的文献求助10
1分钟前
shentaii完成签到,获得积分0
1分钟前
果枳完成签到,获得积分10
1分钟前
行必至完成签到,获得积分20
1分钟前
萧雨墨发布了新的文献求助10
1分钟前
香蕉觅云的应助被1874采纳,获得10
1分钟前
张阳发布了新的文献求助10
1分钟前
萧雨墨发布了新的文献求助10
2分钟前
情怀的应助被Lihedou采纳,获得10
2分钟前
2分钟前
2分钟前
Lihedou发布了新的文献求助10
2分钟前
萧雨墨发布了新的文献求助10
2分钟前
Maestro_S的应助被abc采纳,获得10
2分钟前
2分钟前
Jun的应助被科研通管家采纳,获得10
2分钟前
2分钟前
Jun的应助被科研通管家采纳,获得10
2分钟前
科研通AI65的应助被高速公鹿采纳,获得10
2分钟前
kyokyoro完成签到,获得积分10
2分钟前
在水一方的应助被渟柠采纳,获得20
2分钟前
wsws完成签到 ,获得积分10
2分钟前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 2000
Sex, Aging, and Death in a Medieval Medical Compendium: Trinity College Cambridge MS R.14.52, Its Texts, Language, and Scribe 500
高分子基功能界面水蒸发材料与应用研究 500
外场增强的水汽发电研究 500
基于结构调控的功能化水诱导发电材料与器件研究 500
基于界面调控的高效湿气发电材料与器件研究 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8190117
求助须知:如何正确求助?哪些是违规求助? 9577222
关于积分的说明 21852176
捐赠科研通 7788406
什么是DOI,文献DOI怎么找? 3436259
关于科研通互助平台的介绍 2559994
邀请新用户注册赠送积分活动 2482862