Goal-Oriented Procedure Generation.pdf
The knowledge of scripts, common chains of events in stereotypical scenarios, is a valuable asset for task-oriented natural language understanding systems. We propose the Goal-Oriented Script Construction task, where a model produces a sequence of steps to accomplish a given goal. We pilot our task on the first multilingual script learning dataset supporting 18 languages collected from wikiHow, a website containing half a million how-to articles. For baselines, we consider both a generation-based approach using a language model and a retrieval-based approach by first retrieving the relevant steps from a large candidate pool and then ordering them. We show that our task is practical, feasible but challenging for state-of-the-art Transformer models, and that our methods can be readily deployed for various other datasets and domains with decent zero-shot performance.
The knowledge of scripts, common chains of events in stereotypical scenarios, is a valuable asset for task-oriented natural language understanding systems. We propose the Goal-Oriented Script Construction task, where a model produces a sequence of steps to accomplish a given goal. We pilot our task on the first multilingual script learning dataset supporting 18 languages collected from wikiHow, a website containing half a million how-to articles. For baselines, we consider both a generation-based approach using a language model and a retrieval-based approach by first retrieving the relevant st
Explore this link on the map →related reading
- Natural Language Autoencoders Produce Unsupervised Explanations of LLM Activationstransformer-circuits.pub
- Language Models can Solve Computer Tasksarxiv.org
- GitHub - guidance-ai/guidance: A guidance language for controlling large language models. · GitHubgithub.com
- BIG-bench/bigbench/benchmark_tasks/keywords_to_tasks.md at main · google/BIG-bench · GitHubgithub.com
- [2005.11401] Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasksarxiv.org
- 2305.07759arxiv.org
- TOD-Bertarxiv.org
- 2025.findings-emnlp.313.pdfaclanthology.org
- [2201.07207] Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agentsarxiv.org
- Language Models are Unsupervised Multitask Learnerscdn.openai.com
- GitHub - karpathy/LLM101n: LLM101n: Let's build a Storyteller · GitHubgithub.com
- GitHub - affaan-m/ECC: The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond. · GitHubgithub.com