Language-Conditioned World Modeling for Visual Navigation

• URL: http://arxiv.org/abs/2603.26741v1
• Date: Mon, 23 Mar 2026 04:10:14 GMT
• Title: Language-Conditioned World Modeling for Visual Navigation
• Abstract summary: We study language-conditioned visual navigation (LCVN), in which an embodied agent is asked to follow a natural language instruction.<n>We formulate this problem as open-loop trajectory prediction conditioned on linguistic instructions.<n>We develop LCVN frameworks that link language grounding, future-state prediction, and action generation.
• Score: 48.72354240617526
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We study language-conditioned visual navigation (LCVN), in which an embodied agent is asked to follow a natural language instruction based only on an initial egocentric observation. Without access to goal images, the agent must rely on language to shape its perception and continuous control, making the grounding problem particularly challenging. We formulate this problem as open-loop trajectory prediction conditioned on linguistic instructions and introduce the LCVN Dataset, a benchmark of 39,016 trajectories and 117,048 human-verified instructions that supports reproducible research across a range of environments and instruction styles. Using this dataset, we develop LCVN frameworks that link language grounding, future-state prediction, and action generation through two complementary model families. The first family combines LCVN-WM, a diffusion-based world model, with LCVN-AC, an actor-critic agent trained in the latent space of the world model. The second family, LCVN-Uni, adopts an autoregressive multimodal architecture that predicts both actions and future observations. Experiments show that these families offer different advantages: the former provides more temporally coherent rollouts, whereas the latter generalizes better to unseen environments. Taken together, these observations point to the value of jointly studying language grounding, imagination, and policy learning in a unified task setting, and LCVN provides a concrete basis for further investigation of language-conditioned world models. The code is available at https://github.com/F1y1113/LCVN.

Related papers

• EgoBabyVLM: Benchmarking Cross-Modal Learning from Naturalistic Egocentric Video Data [20.729468513123756]
  Children acquire language grounding with remarkable robustness from limited visuo-linguistic input.<n>Current vision-language models fail to generalize to the sparse, weakly-aligned egocentric streams produced by wearable devices, embodied agents, and infant head-cams.<n>We introduce the EgoBabyVLM Challenge to drive the development of models capable of grounded language learning from the kind of naturalistic data that human infants experience.
  arXiv  Detail & Related papers  (2026-05-18T21:30:54Z)
• Seeing to Act, Prompting to Specify: A Bayesian Factorization of Vision Language Action Policy [59.44168425139687]
  BayesVLA is a Bayesian factorization that decomposes the policy into a visual-action prior, supporting seeing-to-act, and a language-conditioned likelihood, enabling prompt-to-specify.<n>Experiments show superior generalization to unseen instructions, objects, and environments compared to existing methods.
  arXiv  Detail & Related papers  (2025-12-12T01:59:23Z)
• A Novel Cartography-Based Curriculum Learning Method Applied on RoNLI: The First Romanian Natural Language Inference Corpus [71.77214818319054]
  Natural language inference is a proxy for natural language understanding. There is no publicly available NLI corpus for the Romanian language. We introduce the first Romanian NLI corpus (RoNLI) comprising 58K training sentence pairs.
  arXiv  Detail & Related papers  (2024-05-20T08:41:15Z)
• LangNav: Language as a Perceptual Representation for Navigation [63.90602960822604]
  We explore the use of language as a perceptual representation for vision-and-language navigation (VLN) Our approach uses off-the-shelf vision systems for image captioning and object detection to convert an agent's egocentric panoramic view at each time step into natural language descriptions.
  arXiv  Detail & Related papers  (2023-10-11T20:52:30Z)
• Learning to translate by learning to communicate [11.43638897327485]
  We formulate and test a technique to use Emergent Communication (EC) with a pre-trained multilingual model to improve on modern Unsupervised NMT systems. In our approach, we embed a multilingual model into an EC image-reference game, in which the model is incentivized to use multilingual generations to accomplish a vision-grounded task. We present two variants of EC Fine-Tuning (Steinert-Threlkeld et al., 2022), one of which outperforms a backtranslation-only baseline in all four languages investigated.
  arXiv  Detail & Related papers  (2022-07-14T15:58:06Z)
• ZmBART: An Unsupervised Cross-lingual Transfer Framework for Language Generation [4.874780144224057]
  Cross-lingual transfer for natural language generation is relatively understudied. We consider four NLG tasks (text summarization, question generation, news headline generation, and distractor generation) and three syntactically diverse languages. We propose an unsupervised cross-lingual language generation framework (called ZmBART) that does not use any parallel or pseudo-parallel/back-translated data.
  arXiv  Detail & Related papers  (2021-06-03T05:08:01Z)
• Unsupervised Paraphrasing with Pretrained Language Models [85.03373221588707]
  We propose a training pipeline that enables pre-trained language models to generate high-quality paraphrases in an unsupervised setting. Our recipe consists of task-adaptation, self-supervision, and a novel decoding algorithm named Dynamic Blocking. We show with automatic and human evaluations that our approach achieves state-of-the-art performance on both the Quora Question Pair and the ParaNMT datasets.
  arXiv  Detail & Related papers  (2020-10-24T11:55:28Z)
• Generative Language-Grounded Policy in Vision-and-Language Navigation with Bayes' Rule [80.0853069632445]
  Vision-and-language navigation (VLN) is a task in which an agent is embodied in a realistic 3D environment and follows an instruction to reach the goal node. In this paper, we design and investigate a generative language-grounded policy which uses a language model to compute the distribution over all possible instructions. In experiments, we show that the proposed generative approach outperforms the discriminative approach in the Room-2-Room (R2R) and Room-4-Room (R4R) datasets, especially in the unseen environments.
  arXiv  Detail & Related papers  (2020-09-16T16:23:17Z)

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.