paper-with-me

Papers

Causal Inference for Human-Language Model Collaboration

2024-03-30 · Bohan Zhang, Yixin Wang, Paramveer S. Dhillon

In this paper, we examine the collaborative dynamics between humans and language models (LMs), where the interactions typically involve LMs proposing text segments and humans editing or responding to these proposals. Productive engagement with LMs in such scenarios necessitates that humans discern effective text-based interaction strategies, such as editing and response styles, from historical human-LM interactions. This objective is inherently causal, driven by the counterfactual `what-if' question: how would the outcome of collaboration change if humans employed a different text editing/refinement strategy? A key challenge in answering this causal inference question is formulating an appropriate causal estimand: the conventional average treatment effect (ATE) estimand is inapplicable to text-based treatments due to their high dimensionality. To address this concern, we introduce a new causal estimand -- Incremental Stylistic Effect (ISE) -- which characterizes the average impact of infinitesimally shifting a text towards a specific style, such as increasing formality. We establish the conditions for the non-parametric identification of ISE. Building on this, we develop CausalCollab, an algorithm designed to estimate the ISE of various interaction strategies in dynamic human-LM collaborations. Our empirical investigations across three distinct human-LM collaboration scenarios reveal that CausalCollab effectively reduces confounding and significantly improves counterfactual estimation over a set of competitive baselines.

📄 PDF Abstract BibTeX arXiv:2404.00207

Code (1)

xmubq/dtr-text 공식 구현 pytorch

Tasks

Causal InferencecounterfactualLanguage ModelingLanguage Modellingmodel

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically
Causal inference Causal inference is the process of drawing a conclusion about a causal connection based on the conditions of the occurrence of an effect. The main difference between causal…

Similar Papers 제목 키워드 기반

CausalAgent: A Conversational Multi-Agent System for End-to-End Causal Inference

2026-02-12 · Jiawei Zhu, Wei Chen, Ruichu Cai arxiv

Causal inference holds immense value in fields such as healthcare, economics, and social sciences. However, traditional causal analysis workflows impose significant technical barriers, requiring researchers to possess du…

Causal Inference

Large Language Models and Causal Inference in Collaboration: A Comprehensive Survey

2024-03-14 · Xiaoyu Liu, Paiheng Xu, Junda Wu, Jiaxin Yuan 외

Causal inference has shown potential in enhancing the predictive accuracy, fairness, robustness, and explainability of Natural Language Processing (NLP) models by capturing causal relationships among variables. The emerg…

Causal InferenceFairness

Confounding-Robust Policy Improvement with Human-AI Teams

2023-10-13 · Ruijiang Gao, Mingzhang Yin

Human-AI collaboration has the potential to transform various domains by leveraging the complementary strengths of human experts and Artificial Intelligence (AI) systems. However, unobserved confounding can undermine the…

Causal InferenceSensitivity

Evaluation Methods and Measures for Causal Learning Algorithms

2022-02-07 · Lu Cheng, Ruocheng Guo, Raha Moraffah, Paras Sheth 외

The convenient access to copious multi-faceted data has encouraged machine learning researchers to reconsider correlation-based learning and embrace the opportunity of causality-based learning, i.e., causal machine learn…

BenchmarkingBIG-bench Machine LearningCausal Inference

Causal Responsibility Attribution for Human-AI Collaboration

2024-11-05 · Yahang Qi, Bernhard Schölkopf, Zhijing Jin

As Artificial Intelligence (AI) systems increasingly influence decision-making across various fields, the need to attribute responsibility for undesirable outcomes has become essential, though complicated by the complex …

AttributecounterfactualCounterfactual ReasoningDecision Making