aiqbench analysis
This arXiv paper proposes HAIDL, a formal language for representing human-AI interactions as structured, replayable data, along with HAIDL-X, a JupyterLab middleware that implements it. The core claim is that logging and modeling chatbot exchanges as durable records could improve reproducibility, explainability, and behavioral analysis. It is a conceptual and systems proposal with no reported evaluation results in the abstract, so its practical value depends on adoption and on empirical validation of the proposed learning-persona measures.
Human--AI Interactions as First-Class Data: A Formal Interaction Model and Runtime for Explainable AI Collaboration
arXiv:2610.40006v1 Announce Type: new Abstract: AI assistants are increasingly embedded in computational notebooks, scientific workflows, software engineering tools, and learning systems, yet human-AI interactions are often treated as transient prompt-response exchanges rather than durable data products, hindering reproducibility, explainability, replay, behavioral analysis, and adaptive AI support. We argue that these interactions form a new data class requiring formal representation and introduce the Human-AI Interaction Description Language (HAIDL), a formal interaction model with a Backus-Naur Form-inspired concrete syntax. HAIDL encodes each interaction as a structured object capturing computational context, user request, prompt payload, AI response, feedback, problem context, provenance, and model metadata. We integrate this vision into HAIDL-X, a middleware for JupyterLab that instantiates HAIDL by capturing user-bot state, building contextualized prompts, interfacing with AI providers, collecting feedback, and storing interactions in a repository. We add a learning layer over HAIDL that maps interaction objects and notebook execution traces into reproducible measures of user behavior, including engagement, performance, help-seeking, error-oriented use, and learning personas. HAIDL-X provides a foundation for treating human-AI collaboration as structured, queryable, replayable, analyzable data.
Abstract:AI assistants are increasingly embedded in computational notebooks, scientific workflows, software engineering tools, and learning systems, yet human-AI interactions are often treated as transient prompt-response exchanges rather than durable data products, hindering reproducibility, explainability, replay, behavioral analysis, and adaptive AI support. We argue that these interactions form a new data class requiring formal representation and introduce the Human-AI Interaction Description Language (HAIDL), a formal interaction model with a Backus-Naur Form-inspired concrete syntax. HAIDL encodes each interaction as a structured object capturing computational context, user request, prompt payload, AI response, feedback, problem context, provenance, and model metadata. We integrate this vision into HAIDL-X, a middleware for JupyterLab that instantiates HAIDL by capturing user-bot state, building contextualized prompts, interfacing with AI providers, collecting feedback, and storing interactions in a repository. We add a learning layer over HAIDL that maps interaction objects and notebook execution traces into reproducible measures of user behavior, including engagement, performance, help-seeking, error-oriented use, and learning personas. HAIDL-X provides a foundation for treating human-AI collaboration as structured, queryable, replayable, analyzable data.
Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
arXiv:2610.40006 [cs.AI]
(or arXiv:2610.40006v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2610.40006 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 267 words
Related coverage
MOCHI: Motion Enhancement of Collaborative Human-object Interactions
arXiv · Jun 17, 2026
EvoTeam: Multi-Agent Collaboration via Evolving Teams
arXiv · Oct 31, 2026
Analyzing Middle School Students' Dialogue and Behaviors during Collaborative AI Chatbot Development Using Ordered Network Analysis
arXiv · Jul 28, 2026
POI Recommendation with LLM-Augmented Multi-Graph Learning and Contrastive Alignment
arXiv · Aug 19, 2026