Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2605.28465 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914608503062528 |
|---|---|
| author | Park, Jihyeong Baek, Ingeol Park, Jeonghyun Lee, Hwanhee |
| author_facet | Park, Jihyeong Baek, Ingeol Park, Jeonghyun Lee, Hwanhee |
| contents | Divergent thinking is a core dimension of creativity, yet existing evaluations of Large Language Models (LLMs) treat them as single-turn text generations, failing to capture how an agent reasons through iterative interaction. To address this, we introduce MUTATE, an interactive benchmark designed to evaluate agentic divergent thinking at two levels: path-level, where an agent discovers multiple alternative paths to the same goal, and action-level, where individual actions require non-typical, mechanism-shifting object uses. Unlike success-only evaluations, MUTATE scores both completed paths and off-path attempts, capturing divergent reasoning that conventional success rates discard. Our experiments with frontier LLMs reveal a structural blind spot in existing frameworks: when exposed to immediate convergence pressure, they tend to fall into immediate action fixation, failing to improve action-level divergence. To overcome this, we propose ReDNA, which separates unconstrained divergent candidate generation from convergent constraint selection. ReDNA significantly outperforms prior methods across both divergence levels and generalizes effectively to an external creativity environment. We also confirm its success stems from a qualitative enhancement of resilient divergent reasoning rather than simple environmental exploration. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_28465 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Beyond One Path: Evaluating and Enhancing Divergent Thinking in Interactive LLM Agents Park, Jihyeong Baek, Ingeol Park, Jeonghyun Lee, Hwanhee Computation and Language Divergent thinking is a core dimension of creativity, yet existing evaluations of Large Language Models (LLMs) treat them as single-turn text generations, failing to capture how an agent reasons through iterative interaction. To address this, we introduce MUTATE, an interactive benchmark designed to evaluate agentic divergent thinking at two levels: path-level, where an agent discovers multiple alternative paths to the same goal, and action-level, where individual actions require non-typical, mechanism-shifting object uses. Unlike success-only evaluations, MUTATE scores both completed paths and off-path attempts, capturing divergent reasoning that conventional success rates discard. Our experiments with frontier LLMs reveal a structural blind spot in existing frameworks: when exposed to immediate convergence pressure, they tend to fall into immediate action fixation, failing to improve action-level divergence. To overcome this, we propose ReDNA, which separates unconstrained divergent candidate generation from convergent constraint selection. ReDNA significantly outperforms prior methods across both divergence levels and generalizes effectively to an external creativity environment. We also confirm its success stems from a qualitative enhancement of resilient divergent reasoning rather than simple environmental exploration. |
| title | Beyond One Path: Evaluating and Enhancing Divergent Thinking in Interactive LLM Agents |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2605.28465 |