Thoughts without Thinking: Reconsidering the Explanatory Value of Chain-of-Thought Reasoning in LLMs through Agentic Pipelines

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Manuvinakurike, Ramesh, Moss, Emanuel, Watkins, Elizabeth Anne, Sahay, Saurav, Raffa, Giuseppe, Nachman, Lama
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918006846652416
author Manuvinakurike, Ramesh
Moss, Emanuel
Watkins, Elizabeth Anne
Sahay, Saurav
Raffa, Giuseppe
Nachman, Lama
author_facet Manuvinakurike, Ramesh
Moss, Emanuel
Watkins, Elizabeth Anne
Sahay, Saurav
Raffa, Giuseppe
Nachman, Lama
contents Agentic pipelines present novel challenges and opportunities for human-centered explainability. The HCXAI community is still grappling with how best to make the inner workings of LLMs transparent in actionable ways. Agentic pipelines consist of multiple LLMs working in cooperation with minimal human control. In this research paper, we present early findings from an agentic pipeline implementation of a perceptive task guidance system. Through quantitative and qualitative analysis, we analyze how Chain-of-Thought (CoT) reasoning, a common vehicle for explainability in LLMs, operates within agentic pipelines. We demonstrate that CoT reasoning alone does not lead to better outputs, nor does it offer explainability, as it tends to produce explanations without explainability, in that they do not improve the ability of end users to better understand systems or achieve their goals.
format Preprint
id arxiv_https___arxiv_org_abs_2505_00875
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Thoughts without Thinking: Reconsidering the Explanatory Value of Chain-of-Thought Reasoning in LLMs through Agentic Pipelines
Manuvinakurike, Ramesh
Moss, Emanuel
Watkins, Elizabeth Anne
Sahay, Saurav
Raffa, Giuseppe
Nachman, Lama
Artificial Intelligence
Agentic pipelines present novel challenges and opportunities for human-centered explainability. The HCXAI community is still grappling with how best to make the inner workings of LLMs transparent in actionable ways. Agentic pipelines consist of multiple LLMs working in cooperation with minimal human control. In this research paper, we present early findings from an agentic pipeline implementation of a perceptive task guidance system. Through quantitative and qualitative analysis, we analyze how Chain-of-Thought (CoT) reasoning, a common vehicle for explainability in LLMs, operates within agentic pipelines. We demonstrate that CoT reasoning alone does not lead to better outputs, nor does it offer explainability, as it tends to produce explanations without explainability, in that they do not improve the ability of end users to better understand systems or achieve their goals.
title Thoughts without Thinking: Reconsidering the Explanatory Value of Chain-of-Thought Reasoning in LLMs through Agentic Pipelines
topic Artificial Intelligence
url https://arxiv.org/abs/2505.00875