mlsanityData to 7 Oct 2026

Reading and Acting while Blindfolded: The Need for Semantics in Text Game Agents

arXiv 2103.13552 ↗First posted 25 Mar 2021cs.CL cs.AIcited by 20

Not on the map. Why?

Abstract (arXiv)

Text-based games simulate worlds and interact with players using natural language. Recent work has used them as a testbed for autonomous language-understanding agents, with the motivation being that understanding the meanings of words or semantics is a key component of how humans understand, reason, and act in these worlds. However, it remains unclear to what extent artificial agents utilize semantic understanding of the text. To this end, we perform experiments to systematically reduce the amount of semantic information available to a learning agent. Surprisingly, we find that an agent is capable of achieving high scores even in the complete absence of language semantics, indicating that the currently popular experimental setup and models may be poorly designed to understand and leverage game texts. To remedy this deficiency, we propose an inverse dynamics decoder to regularize the representation space and encourage exploration, which shows improved performance on several games including Zork I. We discuss the implications of our findings for designing future agents with stronger semantic understanding.

Key facts. Cited by 20 papers in the map (data through 7 Oct 2026). Builds on 13 of its 15 references that are in the map. First posted on arXiv 25 Mar 2021.

Builds on

13 of 15 references.

References · 13

  1. 2014On the Properties of Neural Machine Translation: Encoder-Decoder Approachescited by 1,303
  2. 2017Curiosity-driven Exploration by Self-supervised Predictioncited by 1,173
  3. 2016Deep Reinforcement Learning for Dialogue Generationcited by 286
  4. 2019Interactive Fiction Games: A Colossal Adventurecited by 165
  5. 2020Keep CALM and Explore: Language Models for Action Generation in Text-based Gamescited by 108
  6. 2015Deep Reinforcement Learning with a Natural Language Action Spacecited by 88
  7. 2020Graph Constrained Reinforcement Learning for Natural Language Action Spacescited by 80
  8. 2015Language Understanding for Text-based Games Using Deep Reinforcement Learningcited by 70
  9. 2020Interactive Fiction Game Playing as Multi-Paragraph Reading Comprehension with Reinforcement Learningcited by 27
  10. 2019NAIL: A General Interactive Fiction Agentcited by 17
  11. 2017What can you do with a rock? Affordance extraction via word embeddingscited by 13
  12. 2020How To Avoid Being Eaten By a Grue: Exploration Strategies for Text-Adventure Agentscited by 10
1 more
  1. 2020Deriving Commonsense Inference Tasks from Interactive Fictionscited by 2

Built on by

Cited by year

Citing papers per year: 2021: 3; 2022: 10; 2023: 5; 2024: 1; 2025: 1.

20 papers.

Sort:

By area:

  1. 2022WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agentscited by 1,197
  2. 2022ScienceWorld: Is your Agent Smarter than a 5th Grader?cited by 298
  3. 2022Pre-Trained Language Models for Interactive Decision-Makingcited by 220
  4. 2023Interactive Natural Language Processingcited by 31
  5. 2023LARP: Language-Agent Role Play for Open-World Gamescited by 24
  6. 2022Multi-Stage Episodic Control for Strategic Exploration in Text Gamescited by 22
  7. 2021Pre-trained Language Models as Prior Knowledge for Playing Text-based Gamescited by 21
  8. 2021A Systematic Survey of Text Worlds as Embodied Natural Language Environmentscited by 17
  9. 2022TVShowGuess: Character Comprehension in Stories as Speaker Guessingcited by 17
  10. 2022TextWorldExpress: Simulating Text Games at One Million Steps Per Secondcited by 12
  11. 2022Behavior Cloned Transformers are Neurosymbolic Reasonerscited by 10
  12. 2023Language Model-In-The-Loop: Data Optimal Approach to Learn-To-Recommend Actions in Text Gamescited by 3
8 more
  1. 2025Learning to Play Like Humans: A Framework for LLM Adaptation in Interactive Fiction Gamescited by 3
  2. 2022Revisiting the Roles of "Text" in Text Gamescited by 2
  3. 2023Learning to Play Text-based Adventure Games with Maximum Entropy Reinforcement Learningcited by 2
  4. 2023Learn What Is Possible, Then Choose What Is Best: Disentangling One-To-Many Relations in Language Through Text-based Gamescited by 2
  5. 2022An Analysis of Deep Reinforcement Learning Agents for Text-based Gamescited by 1
  6. 2021Feudal Reinforcement Learning by Reading Manualscited by 0
  7. 2022JECC: Commonsense Reasoning Tasks Derived from Interactive Fictionscited by 0
  8. 2024On the Effects of Fine-tuning Language Models for Text-Based Reinforcement Learningcited by 0