相关论文: CRAB: Assessing the Strength of Causal Relationshi…

Interpretability research on large language models (LLMs) has yielded important insights into model behaviour, yet recurring pitfalls persist: findings that do not generalise, and causal interpretations that outrun the evidence. Our…

We propose CRAVE (Cluster-based Retrieval Augmented Verification with Explanation); a novel framework that integrates retrieval-augmented Large Language Models (LLMs) with clustering techniques to address fact-checking challenges on social…

Understanding causality in event sequences where outcome labels such as diseases or system failures arise from preceding events like symptoms or error codes is critical. Yet remains an unsolved challenge across domains like healthcare or…

Narratives are fundamental to our perception of the world and are pervasive in all activities that involve the representation of events in time. Yet, modern online information systems do not incorporate narratives in their representation of…

Rumors are rampant in the era of social media. Conversation structures provide valuable clues to differentiate between real and fake claims. However, existing rumor detection methods are either limited to the strict relation of user…

Script event prediction aims to predict the subsequent event given the context. This requires the capability to infer the correlations between events. Recent works have attempted to improve event correlation reasoning by using pretrained…

We introduce CLEAR-3K, a dataset of 3,000 assertion-reasoning questions designed to evaluate whether language models can determine if one statement causally explains another. Each question present an assertion-reason pair and challenge…

People from all over the world use social media to share thoughts and opinions about events, and understanding what people say through these channels has been of increasing interest to researchers, journalists, and marketers alike. However,…

We present a constraint-based algorithm for learning causal structures from observational time-series data, in the presence of latent confounders. We assume a discrete-time, stationary structural vector autoregressive process, with both…

Correct answers do not necessarily reflect cultural understanding. We introduce CRaFT, an explanation-based multilingual evaluation framework designed to assess how large language models (LLMs) reason across cultural contexts. Rather than…

In text classification tasks, models often rely on spurious correlations for predictions, incorrectly associating irrelevant features with the target labels. This issue limits the robustness and generalization of models, especially when…

Causal reasoning is a cornerstone of human intelligence and a critical capability for artificial systems aiming to achieve advanced understanding and decision-making. This thesis delves into various dimensions of causal reasoning and…

In modern advanced emergency management systems many solutions for decision support have been provided as attempts to support humans to take important decisions for the critical situations recovery. The critical situation detection is a…

We introduce CausaLab, a scalable environment for evaluating interactive causal discovery by LLM agents. Unlike prior evaluations, CausaLab evaluates both whether an agent can solve a problem using causal evidence and whether its answer is…

Computational modeling is crucial for understanding and analyzing complex systems. In biology, model creation is a human dependent task that requires reading hundreds of papers and conducting wet lab experiments, which would take days or…

We assess the reliability of relational event model parameters estimated under two sampling schemes: (1) uniform sampling from the observed events and (2) case-control sampling which samples non-events, or null dyads ("controls"), from a…

Evaluating hypothetical statements about how the world would be had a different course of action been taken is arguably one key capability expected from modern AI systems. Counterfactual reasoning underpins discussions in fairness, the…

Large language models (LLMs) acquire most of their knowledge during pretraining, which ties them to a fixed snapshot of the world and makes adaptation to continuously evolving knowledge challenging. As facts, entities, and events change…

Current Reinforcement Learning (RL) methods often suffer from sample-inefficiency, resulting from blind exploration strategies that neglect causal relationships among states, actions, and rewards. Although recent causal approaches aim to…

Recently, there has been an increasing interest in the construction of general-domain and domain-specific causal knowledge graphs. Such knowledge graphs enable reasoning for causal analysis and event prediction, and so have a range of…