Beyond Semantic Similarity: Open Challenges for Embedding-Based Creative Process Analysis Across AI Design Tools
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912952114741248 |
|---|---|
| author | Lee, Seung Won Jin, Semin Hyun, Kyung Hoon |
| author_facet | Lee, Seung Won Jin, Semin Hyun, Kyung Hoon |
| contents | AI-based creativity support tools (CSTs) are evaluated through domain-specific metrics, limiting cross-domain comparison of creative processes. Embedding-based protocol analysis offers a potential domain-agnostic analytical layer. However, we argue that fixed embedding similarity can misrepresent creative dynamics: it may not detect creative pivots that occur within superficially similar language, treating shifts in the problem being addressed as continued elaboration. We identify three open challenges stemming from this gap: aligning similarity measures with creative significance, segmenting and representing multimodal design traces, and evaluating agentic systems where embedding-based metrics enter the generation loop and shape agent behavior. We propose context-aware interventions using large language models as a direction for making trace analysis sensitive to session-specific creative dynamics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_07611 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Beyond Semantic Similarity: Open Challenges for Embedding-Based Creative Process Analysis Across AI Design Tools Lee, Seung Won Jin, Semin Hyun, Kyung Hoon Human-Computer Interaction AI-based creativity support tools (CSTs) are evaluated through domain-specific metrics, limiting cross-domain comparison of creative processes. Embedding-based protocol analysis offers a potential domain-agnostic analytical layer. However, we argue that fixed embedding similarity can misrepresent creative dynamics: it may not detect creative pivots that occur within superficially similar language, treating shifts in the problem being addressed as continued elaboration. We identify three open challenges stemming from this gap: aligning similarity measures with creative significance, segmenting and representing multimodal design traces, and evaluating agentic systems where embedding-based metrics enter the generation loop and shape agent behavior. We propose context-aware interventions using large language models as a direction for making trace analysis sensitive to session-specific creative dynamics. |
| title | Beyond Semantic Similarity: Open Challenges for Embedding-Based Creative Process Analysis Across AI Design Tools |
| topic | Human-Computer Interaction |
| url | https://arxiv.org/abs/2603.07611 |