Research map / Probabilistic and scientific modeling / Molecules and proteins
Plan, Attend, Generate: Planning for Sequence-to-Sequence Models
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 alignment planning with PAG | attention encoder-decoder · Eulerian Circuits (7-node test set) | accuracy: 90.4% -> 100% | Helped |
| add alignment planning with PAG | question-generation encoder-decoder · SQuAD question generation test set | NLL: 66.25 -> 65.45 | Helped |
| add alignment planning with PAG | character-level NMT · WMT’15 En→De Test 2015 | BLEU: 23.45 -> 22.42 | Hurt |
| add alignment planning with PAG | character-level NMT · WMT’15 En→Cs Test 2015 | BLEU: 16.98 -> 18.88 | Helped |
| add alignment planning with PAG | character-level NMT · WMT’15 En→Fi Test 2015 | BLEU: 10.93 -> 11.08 | Helped |
| add layer normalization (ablation) | PAG · WMT’15 En→De Test 2015 | BLEU: 22.42 -> 23.18 | Helped |
| add layer normalization (ablation) | rPAG · WMT’15 En→De Test 2015 | BLEU: 22.85 -> 22.83 | No clear effect |
| add layer normalization (ablation) | PAG · WMT’15 En→Cs Test 2015 | BLEU: 18.88 -> 19.48 | Helped |
| add layer normalization (ablation) | PAG · WMT’15 En→Fi Test 2015 | BLEU: 11.08 -> 12.15 | Helped |
Closest papers by meaning
- Test-Time Alignment for Large Language Models via Textual Model Predictive Control (ICLR 2026 · similarity 0.87)
- Plan-and-Act: Improving Planning of Agents for Long-Horizon Tasks (ICML 2025 · similarity 0.87)
- Joint Optimization of Tool Creation and Use for Large Language Model Agents (NeurIPS 2026 · similarity 0.87)
- Language Control Diffusion: Efficiently Scaling through Space, Time, and Tasks (ICLR 2024 · similarity 0.87)
- Sequence to Sequence Reward Modeling: Improving RLHF by Language Feedback (AAAI 2025 · similarity 0.86)
- Improving Large Language Model Planning with Action Sequence Similarity (ICLR 2025 · similarity 0.86)
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.