Cover Image

Why LLMs Fail at Causal Discovery and How Interventional Agents Escape

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Roy, Amartya, Parbhoo, Sonali
Format:	Preprint
Published:	2026
Subjects:	Artificial Intelligence Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

When Verification Fails: How Compositionally Infeasible Claims Escape Rejection
by: Liu, Muxin, et al.
Published: (2026)

How LLMs Fail to Support Fact-Checking
by: Proma, Adiba Mahbub, et al.
Published: (2025)

Why We Need World Models for AGI: Where LLMs Fail and How World Models May Outperform
by: Alaswad, Feisal, et al.
Published: (2026)

Why Do LLM-based Web Agents Fail? A Hierarchical Planning Perspective
by: Aghzal, Mohamed, et al.
Published: (2026)

Estimating Causal Effects of Text Interventions Leveraging LLMs
by: Guo, Siyi, et al.
Published: (2024)

Look Within, Why LLMs Hallucinate: A Causal Perspective
by: Li, He, et al.
Published: (2024)

Adaptation Odyssey in LLMs: Why Does Additional Pretraining Sometimes Fail to Improve?
by: Öncel, Fırat, et al.
Published: (2024)

Accommodation and Epistemic Vigilance: A Pragmatic Account of Why LLMs Fail to Challenge Harmful Beliefs
by: Cheng, Myra, et al.
Published: (2026)

Why Chain of Thought Fails in Clinical Text Understanding
by: Wu, Jiageng, et al.
Published: (2025)

Why Retrieval-Augmented Generation Fails: A Graph Perspective
by: Guo, Kai, et al.
Published: (2026)

How Reliable are Causal Probing Interventions?
by: Canby, Marc, et al.
Published: (2024)

Larger Language Models Don't Care How You Think: Why Chain-of-Thought Prompting Fails in Subjective Tasks
by: Chochlakis, Georgios, et al.
Published: (2024)

Tracking the Limits of Knowledge Propagation: How LLMs Fail at Multi-Step Reasoning with Conflicting Knowledge
by: Feng, Yiyang, et al.
Published: (2026)

Mathematical Reasoning via Intervention-Based Time-Series Causal Discovery Using LLMs as Concept Mastery Simulators
by: Okita, Tsuyoshi
Published: (2026)

Benchmarking LLMs for Pairwise Causal Discovery in Biomedical and Multi-Domain Contexts
by: Anuyah, Sydney, et al.
Published: (2026)

Surgical Feature-Space Decomposition of LLMs: Why, When and How?
by: Chavan, Arnav, et al.
Published: (2024)

Multi-Agent Causal Discovery Using Large Language Models
by: Le, Hao Duong, et al.
Published: (2024)

Causal Intervention-Based Memory Selection for Long-Horizon LLM Agents
by: Srivastava, Saksham Sahai
Published: (2026)

Mechanistic Interpretability of GPT-2: Lexical and Contextual Layers in Sentiment Analysis
by: Hatua, Amartya
Published: (2025)

Food Noise & False Safety: A Systematic Evaluation of How LLMs Fail to Adapt to Eating Disorder Queries with Clinician Feedback
by: Pucci, Giulia, et al.
Published: (2026)

CauScientist: Teaching LLMs to Respect Data for Causal Discovery
by: Peng, Bo, et al.
Published: (2026)

Lost in the Noise: How Reasoning Models Fail with Contextual Distractors
by: Lee, Seongyun, et al.
Published: (2026)

Why Reasoning Fails to Plan: A Planning-Centric Analysis of Long-Horizon Decision Making in LLM Agents
by: Wang, Zehong, et al.
Published: (2026)

The Invisible Leash: Why RLVR May or May Not Escape Its Origin
by: Wu, Fang, et al.
Published: (2025)

Why are LLMs' abilities emergent?
by: Havlík, Vladimír
Published: (2025)

Why AI-Generated Text Detection Fails: Evidence from Explainable AI Beyond Benchmark Accuracy
by: Pudasaini, Shushanta, et al.
Published: (2026)

CausalDetox: Causal Head Selection and Intervention for Language Model Detoxification
by: Wang, Yian, et al.
Published: (2026)

Detection Is Cheap, Routing Is Learned: Why Refusal-Based Alignment Evaluation Fails
by: Frank, Gregory N.
Published: (2026)

How Emotion Shapes the Behavior of LLMs and Agents: A Mechanistic Study
by: Sun, Moran, et al.
Published: (2026)

Causal Interventions on Causal Paths: Mapping GPT-2's Reasoning From Syntax to Semantics
by: Lee, Isabelle, et al.
Published: (2024)

How Far Are We on the Decision-Making of LLMs? Evaluating LLMs' Gaming Ability in Multi-Agent Environments
by: Huang, Jen-tse, et al.
Published: (2024)

Large Language Models for Constrained-Based Causal Discovery
by: Cohrs, Kai-Hendrik, et al.
Published: (2024)

InteractWeb-Bench: Can Multimodal Agent Escape Blind Execution in Interactive Website Generation?
by: Wang, Qiyao, et al.
Published: (2026)

The Validation Gap: A Mechanistic Analysis of How Language Models Compute Arithmetic but Fail to Validate It
by: Bertolazzi, Leonardo, et al.
Published: (2025)

Improving Causal Interventions in Amnesic Probing with Mean Projection or LEACE
by: Dobrzeniecka, Alicja, et al.
Published: (2025)

Causal Understanding by LLMs: The Role of Uncertainty
by: Lithgow-Serrano, Oscar, et al.
Published: (2025)

LLMs Are Prone to Fallacies in Causal Inference
by: Joshi, Nitish, et al.
Published: (2024)

CausalAbstain: Enhancing Multilingual LLMs with Causal Reasoning for Trustworthy Abstention
by: Sun, Yuxi, et al.
Published: (2025)

CAMO: An Agentic Framework for Automated Causal Discovery from Micro Behaviors to Macro Emergence in LLM Agent Simulations
by: Yu, Xiangning, et al.
Published: (2026)

The Mirage of Performance Gains: Why Contrastive Decoding Fails to Mitigate Object Hallucinations in MLLMs?
by: Yin, Hao, et al.
Published: (2025)