A closed-loop self-evolving training system for spatial reasoning in MLLMs that iteratively generates QA pairs matched to the model's current capabilities via confidence feedback, achieving gains with an order of magnitude less data.
Spacevista: All-scale visual spatial reasoning from mm to km
11 Pith papers cite this work. Polarity classification is still indexing.
11
Pith papers citing it
abstract
With the current surge in spatial reasoning explorations, researchers have made significant progress in understanding indoor scenes, but still struggle with diverse applications such as robotics and autonomous driving. This paper aims to advance all-scale spatial reasoning across diverse scenarios by tackling two key challenges: 1) the heavy reliance on indoor 3D scans and labor-intensive manual annotations for dataset curation; 2) the absence of effective all-scale scene modeling, which often leads to overfitting to individual scenes. In this paper, we introduce a holistic solution that integrates a structured spatial reasoning knowledge system, scale-aware modeling, and a progressive training paradigm, as the first attempt to broaden the all-scale spatial intelligence of MLLMs to the best of our knowledge. Using a task-specific, specialist-driven automated pipeline, we curate over 38K video scenes across 5 spatial scales to create SpaceVista-1M, a dataset comprising approximately 1M spatial QA pairs spanning 19 diverse task types. While specialist models can inject useful domain knowledge, they are not reliable for evaluation. We then build an all-scale benchmark with precise annotations by manually recording, retrieving, and assembling video-based data. However, naive training with SpaceVista-1M often yields suboptimal results due to the potential knowledge conflict. Accordingly, we introduce SpaceVista-7B, a spatial reasoning model that accepts dense inputs beyond semantics and uses scale as an anchor for scale-aware experts and progressive rewards. Finally, extensive evaluations across 5 benchmarks, including our SpaceVista-Bench, demonstrate competitive performance, showcasing strong generalization across all scales and scenarios. Our dataset, model, and benchmark will be released on https://peiwensun2000.github.io/mm2km .
citation-role summary
baseline 1
citation-polarity summary
roles
baseline 1polarities
baseline 1representative citing papers
X-Stream benchmark shows SOTA MLLMs score ~50% on concurrent multi-stream tasks and lack proactive ability, using a dual-verification pipeline to avoid single-stream bias.
SpatialAct benchmark shows VLMs handle isolated spatial reasoning but fail to maintain coherent spatial beliefs and produce reliable actions in multi-turn 3D interactions, underperforming humans.
This survey introduces the Generate-Filter-Control-Replay (GFCR) taxonomy to structure rollout pipelines for RL-based post-training of reasoning LLMs.
CamReasoner uses structured O-T-A reasoning and RL on 56k samples to lift camera movement classification from 73.8% to 78.4% and VQA from 60.9% to 74.5% on Qwen2.5-VL-7B.
4D-RGPT uses perceptual 4D distillation to boost region-level 4D perception in multimodal LLMs and reports gains on existing and new video QA benchmarks.
GASP injects geometric priors into VLMs via a deep-supervised correspondence head trained on video point correspondences and depth consistency, raising internal matching accuracy and delivering gains on spatial benchmarks without any 3D VQA data.
Presents LongSpace-Bench benchmark and LongSpace framework that chunks long videos, adds 3D structural cues, and builds layer-aware memory to improve spatial reasoning in multimodal LLMs.
OneThinker unifies image and video reasoning in one model across 10 tasks via a 600k corpus, CoT-annotated SFT, and EMA-GRPO reinforcement learning, reporting strong results on 31 benchmarks plus some cross-task transfer.
SpaceEra++ adds ScenePick frame sampling and SpaceAlign pairwise constraints to the prior SpaceEra system, claiming consistent benchmark gains for 3D video spatial reasoning.