Agentic Path Maps: Exposing an Agent's Traversal Through the Garden of Forking Paths
cs.HC, cs.AI
Submitted: 2026-11-03
Updated: 2026-11-03
Code: https://github.com/madeleinegrunde/agentic_path_maps
License: http://creativecommons.org/licenses/by/4.0/
The gist: When an AI agent executes an underspecified task, it embeds choices that may contradict unarticulated user context and preferences.
Abstract
When an AI agent executes an underspecified task, it embeds choices that may contradict unarticulated user context and preferences. However, interrupting the user for clarification on all such decisions is untenable. How might agents communicate impactful, but unexplored, paths to the user after execution? To investigate, we first distill a design space for communicating alternative paths, supporting the generation of new systems and structuring dimensions of future study. We then measure the impact of one vital design space dimension on user comprehension: the number of alternatives displayed. Through a user study (n=48) in the data analysis domain, we find that participants rarely consider alternative paths of their own volition, but AI overreliance reduces when alternative paths are visualized for them. By revealing paths the agent did not take, we demonstrate how users can better interpret results and incorporate personal context.
Sources
- Faithfulness vs. Plausibility: On the (Un)Reliability of Explanations from Large Language Models
- AutoDiscovery: Open-ended Scientific Discovery via Bayesian Surprise
- Challenges in Human-Agent Communication
- PLanet: Formalizing and Analyzing Assignment Procedures in the Design of Experiments
- iToT: An Interactive System for Customized Tree-of-Thought Generation
- TRAIL: Trace Reasoning and Agentic Issue Localization
- Overseeing Agents Without Constant Oversight: Challenges and Opportunities
- Measuring and mitigating overreliance to build human-compatible AI
- The Semantic Scholar Open Data Platform
- Approximation and Progressive Display of Multiverse Analyses
- Data-driven Discovery with Large Generative Models
- The Agentic Garden of Forking Paths
- Magentic-UI: Towards Human-in-the-loop Agentic Systems
- Improving Usability of User Centric Decision Making of Multi-Attribute Products on E-commerce Websites
- Interactive Reasoning: Visualizing and Controlling Chain-of-Thought Reasoning in Large Language Models
- Proactive Decision Support using Automated Planning
- Exploration vs. Fixation: Scaffolding Divergent and Convergent Thinking for Human-AI Co-Creation with Generative Models
- Navigating the Conceptual Multiverse
Related papers
- EmoPhone: A Multi-Wave Dataset for In-the-Wild Mobile and Wearable Affect Sensing
- Tasteply: Externalizing Stable and Situational Preference as Manipulable Spatial Objects
- Shielded Trajectory Sampling for Personalized Relational Safety in Conversational AI
- From Tasks to Turns: Measuring Human Labor Contributions in Human-Agent Joint Production
- Reconstruction and Reflection of Positive Experiences through Resurfacing Laughter-indexed Everyday Moments
- LLM-Assisted Reranking to Operationalize Nuanced Objectives in Recommender Systems