Research map / Reinforcement learning / Visual and representation RL
Personalization as Inverse Planning: Learning Latent Design Intents for Agentic Slide Generation via Structural Denoising
Tianci Liu, Zihan Dong, Linjun Zhang, Haoyu Wang, Jing Gao, Emre Kıcıman, Ranveer Chandra, Wei-Ting Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add RL preference-aligned fine-tuning (ablation) | Spire (Qwen2.5-VL-7B) · Slide personalization test pages | VLM-judge AVG: 0.3230 -> 0.5415 | Improved |
| add RL preference-aligned fine-tuning (ablation) | Spire (Qwen2.5-VL-7B) · Slide personalization test pages | Visual similarity AVG: 0.3448 -> 0.7576 | Improved |
| add iterative revision with critic feedback (ablation) | Spire · Slide personalization test pages | VLM-judge AVG: 0.4038 -> 0.5415 | Improved |
| add iterative revision with critic feedback (ablation) | Spire · Slide personalization test pages | Visual similarity AVG: 0.6698 -> 0.7576 | Improved |
| replace o4-mini critic with Spire-trained 7B critic (ablation) | PSP (o4-mini) · Slide personalization test pages | VLM-judge AVG: 0.4784 -> 0.5035 | Improved |
| replace o4-mini critic with Spire-trained 7B critic (ablation) | PSP (o4-mini) · Slide personalization test pages | Visual similarity AVG: 0.8062 -> 0.9000 | Improved |
| add iterative revision with critic feedback (ablation) | Spire · Slide personalization test pages | VLM-judge faithfulness: 0.4865 -> 0.7072 | Improved |
| add RL fine-tuning (ablation) | Spire · Slide personalization test pages | VLM-judge faithfulness: 0.2995 -> 0.7072 | Improved |
| replace o4-mini critic with Spire-trained 7B critic (ablation) | PSP (o4-mini) · Slide personalization test pages | VLM-judge faithfulness: 0.5504 -> 0.5777 | Improved |
Closest papers by meaning
- Presenting a Paper is an Art: Self-Improvement Aesthetic Agents for Academic Presentations (ICLR 2026 · similarity 0.83)
- MaskPLAN: Masked Generative Layout Planning from Partial Input (CVPR 2024 · similarity 0.83)
- Frontier-Eng: Benchmarking Self-Evolving Agents on Real-World Engineering with Generative Optimization (NeurIPS 2026 · similarity 0.82)
- ScaleEnv: Scaling Environment Synthesis from Scratch for Generalist Interactive Tool-Use Agent Training (ICML 2026 · similarity 0.82)
- AutoPresent: Designing Structured Visuals from Scratch (CVPR 2025 · similarity 0.82)
- ReGen: Generative Robot Simulation via Inverse Design (ICLR 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
