CHI 2025 · tier3_two_buckets_with_human_or_agent
Data Formulator 2: Iterative Creation of Data Visualizations, with AI Transforming Data Along the Way
Directly targets iterative human-AI data visualization authoring where users specify intent and delegate data transformation to AI in exploratory analysis workflows.
ICLR.cc 2026 · tier3_two_buckets_with_human_or_agent
AI-for-Science Low-code Platform with Bayesian Adversarial Multi-Agent Framework
This low-code AI-for-science platform explicitly coordinates multiple LLM agents with human-AI collaboration for scientific code generation and non-expert user workflows.
ICML.cc 2025 · tier3_two_buckets_with_human_or_agent
Are Large Language Models Ready for Multi-Turn Tabular Data Analysis?
Directly targets human-machine conversational tabular data analysis with multi-turn interaction and practical scenarios, aligning strongly with interactive data science workflows.
CHI 2024 · tier3_two_buckets_with_human_or_agent
Marco: Supporting Business Document Workflows via Collection-Centric Information Foraging with Large Language Models
A mixed-initiative LLM workspace for real-world document collection sensemaking with interactive user-AI schema and information foraging support.
CHI 2025 · tier3_two_buckets_with_human_or_agent
ChainBuddy: An AI-assisted Agent System for Generating LLM Pipelines
A user-facing AI workflow assistant for generating and evaluating LLM pipelines with attention to expertise, workload, and over-reliance.
CHI 2025 · tier3_two_buckets_with_human_or_agent
Do It For Me vs. Do It With Me: Investigating User Perceptions of Different Paradigms of Automation in Copilots for Feature-Rich Software
Directly compares automation paradigms and user control in LLM copilots across data analysis and design tasks, aligning with division of labor in interactive software workflows.
CHI 2025 · tier3_two_buckets_with_human_or_agent
Generative AI in Knowledge Work: Design Implications for Data Navigation and Decision-Making
Directly addresses GenAI support for knowledge work involving data navigation, synthesis, decision-making, user control, transparency, and workflow-adaptive collaboration.
CHI 2026 · tier3_two_buckets_with_human_or_agent
Augmenting Clinical Decision-Making with an Interactive and Interpretable AI Copilot: A Real-World User Study with Clinicians in Nephrology and Obstetrics
An interactive AI copilot for clinicians using longitudinal EHR data with explicit verification, expertise differences, and collaborative decision-making in a real-world workflow.
CHI 2026 · tier3_two_buckets_with_human_or_agent
Cocoa: Co-Planning and Co-Execution with AI Agents
Directly addresses human-agent co-planning, co-execution, delegation, and interleaved execution for complex research tasks with notebook-inspired workflows.
CHI 2026 · tier3_two_buckets_with_human_or_agent
DiagLink: A Dual-User Diagnostic Assistance System by Synergizing Experts with LLMs and Knowledge Graphs
A dual-user LLM and knowledge-graph diagnostic system with guided dialogue, physician oversight, role-adaptive interfaces, and collaborative reasoning in a real workflow.
ICLR.cc 2026 · tier3_two_buckets_with_human_or_agent
Collaborative Gym: A Framework for Enabling and Evaluating Human-Agent Collaboration
Directly matches the anchor with bidirectional human-agent collaboration, flexible coordination, real-user evaluation, and a tabular data analysis task.
ACL 2025 · tier3_two_buckets_with_human_or_agent
Controllable and Reliable Knowledge-Intensive Task-Oriented Conversational Agents with Declarative Genie Worksheets
Genie centrally addresses controllable task-oriented conversational agents for complex user interactions with policy and state-management mechanisms plus a user study, though not specifically data/graph focused.
ACL 2026 · tier3_two_buckets_with_human_or_agent
From Individual Excellence to Collective Sustainability: Seeking Strategic Equilibrium in Proactive Multi-Agent Teams
This explicitly studies proactive agents in human-AI scientific teams, task allocation, load balancing, and sustainable collaboration, though data/graph analysis is only tangential.
CHI 2025 · tier3_two_buckets_with_human_or_agent
Beyond Code Generation: LLM-supported Exploration of the Program Design Space
An LLM-supported IDE for iterative program design explicitly studies programmer-agent interaction, design decisions, and information handoff, adjacent to data-science coding workflows.
CHI 2025 · tier3_two_buckets_with_human_or_agent
Interactive Debugging and Steering of Multi-Agent AI Systems
Directly studies interfaces for developers to debug and steer multi-agent AI systems in complex agentic workflows, despite limited data or graph specificity.