Evaluating the World Model Implicit in a Generative Model

• URL: http://arxiv.org/abs/2406.03689v2
• Date: Sat, 22 Jun 2024 18:23:08 GMT
• Title: Evaluating the World Model Implicit in a Generative Model
• Authors: Keyon Vafa, Justin Y. Chen, Jon Kleinberg, Sendhil Mullainathan, Ashesh Rambachan,
• Abstract summary: Recent work suggests that large language models may implicitly learn world models. This includes problems as diverse as simple logical reasoning, geographic navigation, game-playing, and chemistry. We propose new evaluation metrics for world model recovery inspired by the classic Myhill-Nerode theorem from language theory.
• Score: 7.317896355747284
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent work suggests that large language models may implicitly learn world models. How should we assess this possibility? We formalize this question for the case where the underlying reality is governed by a deterministic finite automaton. This includes problems as diverse as simple logical reasoning, geographic navigation, game-playing, and chemistry. We propose new evaluation metrics for world model recovery inspired by the classic Myhill-Nerode theorem from language theory. We illustrate their utility in three domains: game playing, logic puzzles, and navigation. In all domains, the generative models we consider do well on existing diagnostics for assessing world models, but our evaluation metrics reveal their world models to be far less coherent than they appear. Such incoherence creates fragility: using a generative model to solve related but subtly different tasks can lead it to fail badly. Building generative models that meaningfully capture the underlying logic of the domains they model would be immensely valuable; our results suggest new ways to assess how close a given model is to that goal.

Related papers

• Meta-Models: An Architecture for Decoding LLM Behaviors Through Interpreted Embeddings and Natural Language [0.0]
  We use a "meta-model" that takes activations from an "input-model" and answers natural language questions about the input-model's behaviors. We evaluate the meta-model's ability to generalize by training them on selected task types and assessing their out-of-distribution performance in deceptive scenarios.
  arXiv  Detail & Related papers  (2024-10-03T13:25:15Z)
• Promises and Pitfalls of Generative Masked Language Modeling: Theoretical Framework and Practical Guidelines [74.42485647685272]
  We focus on Generative Masked Language Models (GMLMs) We train a model to fit conditional probabilities of the data distribution via masking, which are subsequently used as inputs to a Markov Chain to draw samples from the model. We adapt the T5 model for iteratively-refined parallel decoding, achieving 2-3x speedup in machine translation with minimal sacrifice in quality.
  arXiv  Detail & Related papers  (2024-07-22T18:00:00Z)
• Elements of World Knowledge (EWOK): A cognition-inspired framework for evaluating basic world knowledge in language models [42.48862540545121]
  We present Elements of World Knowledge (EWOK), a framework for evaluating world modeling in language models. EWOK targets specific concepts from multiple knowledge domains known to be vital for world modeling in humans. We then introduce EWOK-CORE-1.0, a dataset of 4,374 items covering 11 world knowledge domains.
  arXiv  Detail & Related papers  (2024-05-15T17:19:42Z)
• Is Sora a World Simulator? A Comprehensive Survey on General World Models and Beyond [101.15395503285804]
  General world models represent a crucial pathway toward achieving Artificial General Intelligence (AGI) In this survey, we embark on a comprehensive exploration of the latest advancements in world models. We examine challenges and limitations of world models, and discuss their potential future directions.
  arXiv  Detail & Related papers  (2024-05-06T14:37:07Z)
• Automated Statistical Model Discovery with Language Models [34.03743547761152]
  We introduce a method for language model driven automated statistical model discovery. We cast our automated procedure within the principled framework of Box's Loop. Our results highlight the promise of LM-driven model discovery.
  arXiv  Detail & Related papers  (2024-02-27T20:33:22Z)
• Open World Object Detection in the Era of Foundation Models [53.683963161370585]
  We introduce a new benchmark that includes five real-world application-driven datasets. We introduce a novel method, Foundation Object detection Model for the Open world, or FOMO, which identifies unknown objects based on their shared attributes with the base known objects.
  arXiv  Detail & Related papers  (2023-12-10T03:56:06Z)
• Advancing State of the Art in Language Modeling [0.0]
  Generalization is arguably the most important goal of statistical language modeling research. Publicly available benchmarks and papers published with an open-source code have been critical to advancing the field. We propose a simple framework that should help advance the state of the art in language modeling in terms of generalization.
  arXiv  Detail & Related papers  (2023-11-28T12:30:43Z)
• Grounded Decoding: Guiding Text Generation with Grounded Models for Embodied Agents [111.15288256221764]
  Grounded-decoding project aims to solve complex, long-horizon tasks in a robotic setting by leveraging the knowledge of both models. We frame this as a problem similar to probabilistic filtering: decode a sequence that both has high probability under the language model and high probability under a set of grounded model objectives. We demonstrate how such grounded models can be obtained across three simulation and real-world domains, and that the proposed decoding strategy is able to solve complex, long-horizon tasks in a robotic setting by leveraging the knowledge of both models.
  arXiv  Detail & Related papers  (2023-03-01T22:58:50Z)
• Evaluation of Categorical Generative Models -- Bridging the Gap Between Real and Synthetic Data [18.142397311464343]
  We introduce an appropriately scalable evaluation method for generative models. We consider increasingly large probability spaces, which correspond to increasingly difficult modeling tasks. We validate our evaluation procedure with synthetic experiments on both synthetic generative models and current state-of-the-art categorical generative models.
  arXiv  Detail & Related papers  (2022-10-28T21:05:25Z)
• Synthetic Model Combination: An Instance-wise Approach to Unsupervised Ensemble Learning [92.89846887298852]
  Consider making a prediction over new test data without any opportunity to learn from a training set of labelled data. Give access to a set of expert models and their predictions alongside some limited information about the dataset used to train them.
  arXiv  Detail & Related papers  (2022-10-11T10:20:31Z)
• Plausible Counterfactuals: Auditing Deep Learning Classifiers with Realistic Adversarial Examples [84.8370546614042]
  Black-box nature of Deep Learning models has posed unanswered questions about what they learn from data. Generative Adversarial Network (GAN) and multi-objectives are used to furnish a plausible attack to the audited model. Its utility is showcased within a human face classification task, unveiling the enormous potential of the proposed framework.
  arXiv  Detail & Related papers  (2020-03-25T11:08:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.