Information and Updates

Overview

Recent advances in large generative models have turned AI agents into everyday companions. Millions of people now rely on them across domains from design and communication to science and education. These advances offer unprecedented opportunities to support people in open-ended domains by providing a medium for brainstorming ideas, exploring design choices, and thereby improving their creative outcomes. However, existing models and interfaces are primarily designed for automation, and their increasing usage for creative tasks poses new challenges, such as the risk of idea homogeneity and concerns about content copyright. The workshop aims to investigate these opportunities and challenges by focusing the discussions along two thrusts:

For us to fully realize these opportunities and tackle these challenges, it is crucial to build a community comprising researchers, industry professionals, and practitioners that are ``multilingual'' with (a) technical expertise in the cutting-edge advances in generative AI, (b) know-how of designing/deploying generative AI systems for human users, and (c) first-hand experience of working with these systems for creative content creation in everyday work. The goal of this workshop is to foster such a multilingual community. More concretely, the workshop will bring together speakers and participants with diverse backgrounds: (i) researchers in machine learning, generative AI, natural language processing, human-computer interaction, industrial design, social robotics, and human psychology; (ii) industry professionals working on end-user systems such as Midjourney; and (iii) practitioners leveraging generative AI in art and creative industries. Moreover, the workshop program, including talks by diverse speakers and a panel, is designed to break new ground. By fostering collaboration between different communities and stakeholders, we aim to facilitate the development of next-generation technologies that enhance human-AI co-creativity. The workshop will cover several topics of interests, including (but not limited to):

Schedule

You can access the latest schedule on the ICML website. For poster session, we have provided poster board assignments at the following link: Poster-Board-Assignment.txt (poster boards are located in Hall A, near entrance A1). The lightning talks are scheduled in four batches as follows (with Paper IDs to be presented in each batch):

• Lightning Talks A (Paper IDs: 1, 6, 7, 9, 11, 13, 16, 17, 19)
• Lightning Talks B (Paper IDs: 21, 22, 24, 29, 30, 31, 33, 36, 38)
• Lightning Talks C (Paper IDs: 39, 40, 43, 44, 46, 49, 51, 53, 54, 55)
• Lightning Talks D (Paper IDs: 57, 60, 62, 63, 69, 70, 77, 80, 81)

Invited Speakers and Panelists

The workshop will have 6 talks from the invited speakers; additionally, we will have a panel discussion with the speakers.

• Luba Elliott. Art and AI Curator (London, UK).
• Hwajung Hong. KAIST (Daejeon, South Korea).
• John Joon Young Chung. Midjourney (San Francisco, CA, USA).
• Pronita Mehrotra. MindAntix (Redmond, WA, USA).
• Lonneke van der Plas. USI (Lugano, Switzerland).
• Ben Y. Zhao. University of Chicago (Chicago, IL, USA).

Papers

• Paper 1: [Paper] [Poster] Caesar: Deep Agentic Web Exploration for Creative Answer Synthesis.
  
• Paper 3: [Paper] [Poster] Beneath the Surface: Investigating LLMs' Capabilities for Communicating with Subtext.
  
• Paper 4: [Paper] [Poster] Multi-Genre Collection: Scaling Creative Writing Beyond Story-Centric Data.
  
• Paper 6: [Paper] [Poster] CresOWLve: Benchmarking Creative Problem-Solving Over Real-World Knowledge.
  
• Paper 7: [Paper] [Poster] Structured Creativity Methods for Multi-Agent LLMs: Brainwriting Outperforms Disney and Double Diamond on LLM-Judged Originality.
  
• Paper 9: [Paper] [Poster] The Missing Drive: Functional Analogs of Intrinsic Motivation in Large Language Models for Creative Tasks.
  
• Paper 11: [Paper] [Poster] Adrasteia: A Human-in-the-Loop Agentic Workflow for Theory-Oriented Research in Economics.
  
• Paper 12: [Paper] [Poster] CreativityNeuro: Steering Language Model Weights to Improve Divergent Thinking and Reduce Mode Collapse.
  
• Paper 13: [Paper] [Poster] Seeking the Unfamiliar but Memorable: Conceptual Creativity as Meta-Learning.
  
• Paper 16: [Paper] [Poster] A Typography Benchmark for Co-Creative Graphic-Design Agents.
  
• Paper 17: [Paper] [Poster] Optimizing Diversity and Quality through Base-Aligned Model Collaboration.
  
• Paper 19: [Paper] [Poster] Incentives shape how humans co-create with generative AI.
  
• Paper 21: [Paper] [Poster] "I've Seen How This Goes": Characterizing the Diversity of LLM Generations and Human Writing via Progressive Conditional Surprise.
  
• Paper 22: [Paper] [Poster] AutoFiction: Measuring AI ability to execute long-horizon writing tasks.
  
• Paper 24: [Paper] [Poster] Feedback-to-Rubrics: Can We Extract Expert Criteria from Inline Comments?
  
• Paper 26: [Paper] [Poster] Do Semantic Distance Tests Actually Predict Creativity in LLMs?
  
• Paper 29: [Paper] [Poster] Bonsai: Cultivating Author Intent in LLM-Based Interactive Digital Narratives.
  
• Paper 30: [Paper] [Poster] SEA: Understanding Sketch Abstraction Efficiency via Element-Level Commonsense for Human-AI Sketching.
  
• Paper 31: [Paper] [Poster] Trajectory as Creative Intent: Rethinking Drag-Based Interaction in Generative Visual Editing.
  
• Paper 32: [Paper] [Poster] Can Large Language Models Match Human Diversity in Educational Content Generation?
  
• Paper 33: [Paper] [Poster] Object-Level Access to SVG Scenes: A Benchmark and Raster-Grounded Method for Text-Guided Vector Object Extraction.
  
• Paper 36: [Paper] [Poster] Visual Aesthetic Benchmark: Can Frontier Models Judge Beauty?
  
• Paper 38: [Paper] [Poster] Bridging Creative Intent and Visual Quality: Creator-Driven Recurrent Video Generation with Agentic Feedback Loops.
  
• Paper 39: [Paper] [Poster] IDEAFix: Evaluation Framework for Creative Defixation Prompting in LLMs.
  
• Paper 40: [Paper] [Poster] VendiEvolve: Towards Diversity-Aware LLM-Guided Evolutionary Search.
  
• Paper 43: [Paper] [Poster] SVG-BERT: Code-Native Representations for Structure-Aware Vector Graphics Workflows.
  
• Paper 44: [Paper] [Poster] SlideMatcher: Agentic Slide Editing under Item Mismatch.
  
• Paper 46: [Paper] [Poster] When Machines Decide If a Human Wrote It: Creativity in the Age of AI Detectors.
  
• Paper 49: [Paper] [Poster] Creativity from Friction: Human-AI Interaction for Exploratory Structural Design.
  
• Paper 51: [Paper] [Poster] JudgeBreak: Is Taste Hidden in the Layers? Stress-Testing Reward Models and LLM Judges with Representation-Based Subjective Evaluation.
  
• Paper 53: [Paper] [Poster] User Engagement With Intermediate Representations in LLM-Assisted Audio Programming.
  
• Paper 54: [Paper] [Poster] Rethinking Post-training Diversity Collapse: Is Diversity-preserving Post-training Enough?
  
• Paper 55: [Paper] [Poster] Evaluating Design Video Generation: Metrics for Compositional Fidelity.
  
• Paper 57: [Paper] [Poster] GameDevBench: Evaluating Agentic Capabilities Through Game Development.
  
• Paper 60: [Paper] [Poster] NarrativeWorldBench: A Frontier-Saturated Benchmark and a Latent World Model for Long-Horizon Co-Creative Audio Drama.
  
• Paper 62: [Paper] [Poster] We Built a Salon, We Called It Alignment: Aesthetic Reward and the Latent Avant-Garde in Generative Models.
  
• Paper 63: [Paper] [Poster] Süssmayr's Heirs: A Composition-Invariant View of Generative Stylistic Completion.
  
• Paper 67: [Paper] [Poster] On the Transfer of Output Diversity via Synthetic Data in Language Models.
  
• Paper 69: [Paper] [Poster] The Human Creativity Benchmark: Studying Convergence and Divergence in an Expert-Labeled Benchmark.
  
• Paper 70: [Paper] [Poster] "I Didn't Make the Micro Decisions": Measuring, Inducing, and Exposing Goal-Level AI Contributions in Collaboration.
  
• Paper 71: [Paper] [Poster] ArtMine: Discovering and Formalizing Artistic Processes.
  
• Paper 72: [Paper] [Poster] Forty Shades of Blue: Quality-Diversity Alignment via Mode-Conditioned Reinforcement Learning.
  
• Paper 73: [Paper] [Poster] Searching for Synergy in Shared Workspace Human-AI Collaboration.
  
• Paper 77: [Paper] [Poster] Repairing Graphic Design Layouts After Content, Asset, and Canvas Updates: A Source-Conditioned Benchmark.
  
• Paper 78: [Paper] [Poster] Creative Collision: Directorial Persona Steering and Competition in Large Language Models.
  
• Paper 80: [Paper] [Poster] Scalable and Interpretable Authorship Attribution for AI Generated Code.
  
• Paper 81: [Paper] [Poster] Toward User-Orchestrated Human-Multi-Agent Teams for Co-Creation.
  

Organizers

• Adish Singla. Max Planck Institute for Software Systems (Saarbrucken, Germany).
• Abhilasha Ravichander. Max Planck Institute for Software Systems (Kaiserslautern, Germany).
• Liwei Jiang. University of Washington (Seattle, WA, USA).
• Alexander Spangher. Stanford University (Stanford, CA, USA).
• Alice Oh. KAIST (Daejeon, South Korea).

• Jiho Jin. KAIST (Daejeon, South Korea).
• Jun Seong Kim. KAIST (Daejeon, South Korea).
• Changyoon Lee. KAIST (Daejeon, South Korea).
• Manh Hung Nguyen. Max Planck Institute for Software Systems (Saarbrucken, Germany).
• Chao Wen. Max Planck Institute for Software Systems (Saarbrucken, Germany).

Call for Papers

Details about paper submissions are provided below:

The workshop welcomes participation from individuals who do not have something they'd like to submit but are interested in the workshop topics. The workshop aims to facilitate new connections, inspire novel ideas, and create fruitful partnerships.