# Be Consistent! Improving Procedural Text Comprehension using Label Consistency

arXiv: https://arxiv.org/abs/1906.08942 · HTML: https://mlsanity.com/paper/1906.08942/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Jun 2019.

## Built on by (most cited first, 10 of 10)

- [Measuring and Improving Consistency in Pretrained Language Models](https://mlsanity.com/paper/2102.01017/) (2021)
- [The Trickle-down Impact of Reward (In-)consistency on RLHF](https://mlsanity.com/paper/2309.16155/) (2023)
- [Causal Reasoning of Entities and Events in Procedural Texts](https://mlsanity.com/paper/2301.10896/) (2023)
- [OpenPI2.0: An Improved Dataset for Entity Tracking in Texts](https://mlsanity.com/paper/2305.14603/) (2023)
- [LEMON: Language-Based Environment Manipulation via Execution-Guided Pre-training](https://mlsanity.com/paper/2201.08081/) (2022)
- [Hire Me or Not? Examining Language Model's Behavior with Occupation Attributes](https://mlsanity.com/paper/2405.06687/) (2024)
- [Enriching a Model's Notion of Belief using a Persistent Memory](https://mlsanity.com/paper/2104.08401/) (2021)
- [Learning Action Conditions from Instructional Manuals for Instruction Understanding](https://mlsanity.com/paper/2205.12420/) (2022)
- [50 Ways to Bake a Cookie: Mapping the Landscape of Procedural Texts](https://mlsanity.com/paper/2210.17235/) (2022)
- [Procedural Text Understanding via Scene-Wise Evolution](https://mlsanity.com/paper/2203.07600/) (2022)

## Abstract (arXiv)

Our goal is procedural text comprehension, namely tracking how the properties of entities (e.g., their location) change with time given a procedural text (e.g., a paragraph about photosynthesis, a recipe). This task is challenging as the world is changing throughout the text, and despite recent advances, current systems still struggle with this task. Our approach is to leverage the fact that, for many procedural texts, multiple independent descriptions are readily available, and that predictions from them should be consistent (label consistency). We present a new learning framework that leverages label consistency during training, allowing consistency bias to be built into the model. Evaluation on a standard benchmark dataset for procedural text, ProPara (Dalvi et al., 2018), shows that our approach significantly improves prediction performance (F1) over prior state-of-the-art systems.