# A Progressive Training Strategy for Embodied Vision-Language Models to Mitigate Spatio-Temporal Hallucinations

arXiv: https://arxiv.org/abs/2604.10506 · HTML: https://mlsanity.com/paper/2604.10506/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Hallucination detection and mitigation, one of 17 directions in LLM reasoning.
- Builds on 21 of its 43 references that are in the map.
- First posted on arXiv 12 Apr 2026.

## Placed in

- LLM reasoning / [Hallucination detection and mitigation](https://mlsanity.com/lr/hallucination-detection-and-mitigation/)
- Multimodal LLMs / vision-language models / [Hallucination mitigation in LVLMs](https://mlsanity.com/mllm/hallucination-mitigation-in-lvlms/)

## Builds on (most cited first, 21 of 21)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [Decoupled Weight Decay Regularization](https://mlsanity.com/paper/1711.05101/) (2017)
- [Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour](https://mlsanity.com/paper/1706.02677/) (2017)
- [Octo: An Open-Source Generalist Robot Policy](https://mlsanity.com/paper/2405.12213/) (2024)
- [Matterport3D: Learning from RGB-D Data in Indoor Environments](https://mlsanity.com/paper/1709.06158/) (2017)
- [AI2-THOR: An Interactive 3D Environment for Visual AI](https://mlsanity.com/paper/1712.05474/) (2017)
- [Multimodal Chain-of-Thought Reasoning in Language Models](https://mlsanity.com/paper/2302.00923/) (2023)
- [PaliGemma: A versatile 3B VLM for transfer](https://mlsanity.com/paper/2407.07726/) (2024)
- [DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding](https://mlsanity.com/paper/2412.10302/) (2024)
- [AgiBot World Colosseo: A Large-scale Manipulation Platform for Scalable and Intelligent Embodied Systems](https://mlsanity.com/paper/2503.06669/) (2025)
- [VIP: Towards Universal Visual Reward and Representation via Value-Implicit Pre-Training](https://mlsanity.com/paper/2210.00030/) (2022)
- [Goal-Conditioned Imitation Learning using Score-based Diffusion Policies](https://mlsanity.com/paper/2304.02532/) (2023)
- [A Vision-Language-Action-Critic Model for Robotic Real-World Reinforcement Learning](https://mlsanity.com/paper/2509.15937/) (2025)
- [Visual Chain of Thought: Bridging Logical Gaps with Multimodal Infillings](https://mlsanity.com/paper/2305.02317/) (2023)
- [Vision-Language Models as a Source of Rewards](https://mlsanity.com/paper/2312.09187/) (2023)

## Built on by (most cited first, 4 of 4)

- [Bridging the Pose-Semantic Gap: A Cascade Framework for Text-Based Person Anomaly Search](https://mlsanity.com/paper/2604.23282/) (2026)
- [Bridging Physical Reasoning and Task Generalization via Visual Action Outcome Reasoning Alignment](https://mlsanity.com/paper/2607.06522/) (2026)
- [TurboT2VA: Fast Large-Scale Text-to-Video-Audio Generation via Score-Regularized Consistency Distillation](https://mlsanity.com/paper/2608.24674/) (2026)
- [EgoTSR++: Egocentric Spatiotemporal Reasoning for Task Progress Understanding](https://mlsanity.com/paper/2609.31751/) (2026)

## Abstract (arXiv)

Vision-Language Models (VLMs) have made significant strides in static image understanding but continue to face critical hurdles in spatiotemporal reasoning. A major bottleneck is "multi-image reasoning hallucination", where a large performance drop between forward and reverse temporal queries reveals a dependence on superficial shortcuts instead of state-based understanding. To mitigate this, we first develop a new Chain-of-Thought (CoT) dataset that decomposes intricate reasoning into detailed spatiotemporal steps and definitive judgments. Building on this, we present a progressive training framework: it initiates with supervised pre-training on our CoT dataset to instill logical structures, followed by fine-tuning with scalable weakly-labeled data for broader generalization. Our experiments demonstrate that this approach not only improves backbone accuracy but also reduces the forward-backward performance gap from over 70% to only 6.53%. This shows that the method strengthens dynamic reasoning and reduces the inherent temporal biases of current VLMs.