Pith. sign in

REVIEW 10 cited by

Improved Visual-Spatial Reasoning via R1-Zero-Like Training

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2504.00883 v2 pith:QBUSGPPT submitted 2025-04-01 cs.CV cs.AI

Zhenyi Liao , Qingsong Xie , Yanhao Zhang , Zijian Kong , Haonan Lu , Zhenyu Yang , Zhijie Deng This is my paper
classification cs.CVcs.AI
keywords reasoningvisual-spatialmodelmllmstrainingcapacitiesdatasetfine-tuned
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Increasing attention has been placed on improving the reasoning capacities of multi-modal large language models (MLLMs). As the cornerstone for AI agents that function in the physical realm, video-based visual-spatial intelligence (VSI) emerges as one of the most pivotal reasoning capabilities of MLLMs. This work conducts a first, in-depth study on improving the visual-spatial reasoning of MLLMs via R1-Zero-like training. Technically, we first identify that the visual-spatial reasoning capacities of small- to medium-sized Qwen2-VL models cannot be activated via Chain of Thought (CoT) prompts. We then incorporate GRPO training for improved visual-spatial reasoning, using the carefully curated VSI-100k dataset, following DeepSeek-R1-Zero. During the investigation, we identify the necessity to keep the KL penalty (even with a small value) in GRPO. With just 120 GPU hours, our vsGRPO-2B model, fine-tuned from Qwen2-VL-2B, can outperform the base model by 12.1% and surpass GPT-4o. Moreover, our vsGRPO-7B model, fine-tuned from Qwen2-VL-7B, achieves performance comparable to that of the best open-source model LLaVA-NeXT-Video-72B. Additionally, we compare vsGRPO to supervised fine-tuning and direct preference optimization baselines and observe strong performance superiority. The code and dataset will be available soon.

Discussion (0). Sign in to comment.

Forward citations

Cited by 10 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Beyond Single Expert: Harmonizing Diverse Visual Priors in MLLMs for Spatial Understanding

    cs.CV 2026-07 conditional novelty 6.0 of 10

    ViPS fuses five complementary visual priors into an MLLM via lightweight distillation proxies and query-conditioned dynamic weighting, reporting state-of-the-art results on VSI-Bench and ScanNet-series benchmarks.

  2. EgoMind: Activating Spatial Cognition through Linguistic Reasoning in MLLMs

    cs.CV 2026-04 unverdicted novelty 6.0 of 10

    EgoMind uses Role-Play Caption and Progressive Spatial Analysis to give MLLMs competitive multi-frame spatial reasoning without 3D priors, using only 5K SFT and 20K RL samples.

  3. Generation Models Know Space: Unleashing Implicit 3D Priors for Scene Understanding

    cs.CV 2026-03 unverdicted novelty 6.0 of 10

    VEGA-3D extracts intermediate spatiotemporal features from a pretrained video diffusion model and fuses them into MLLMs to improve geometric and embodied reasoning without explicit 3D supervision.

  4. From Correspondence to Actions: Human-Like Multi-Image Spatial Reasoning in Multi-modal Large Language Models

    cs.CV 2026-02 conditional novelty 6.0 of 10

    A 3B multimodal LLM trained with patch-level cross-view alignment plus explicit viewpoint-action reasoning outperforms much larger models on two multi-image spatial reasoning benchmarks.

  5. VKnowU: Evaluating Visual Knowledge Understanding in Multimodal LLMs

    cs.CV 2025-11 conditional novelty 6.0 of 10

    A 1,680-question video benchmark shows leading multimodal models lag humans by ~15 points on visual knowledge, and a See-Think-Answer RL-trained model narrows the gap.

  6. EgoExo-Con: Exploring View-Invariant Video Temporal Understanding

    cs.CV 2025-10 conditional novelty 6.0 of 10

    Most Video-LLMs answer temporal questions far less consistently when the same event is shown from ego and exo views, and a GRPO variant with a reasoning-similarity reward partially closes the gap.

  7. Why Do MLLMs Struggle with Spatial Understanding? A Systematic Analysis from Data to Architecture

    cs.CV 2025-09 conditional novelty 6.0 of 10

    Spatial understanding in multimodal LLMs plateaus quickly as training data grows, and position encoding in the visual encoder is the more influential factor.

  8. WeThink: Toward General-purpose Vision-Language Reasoning via Reinforcement Learning

    cs.CV 2025-06 conditional novelty 5.0 of 10

    WeThink, a 120K-image QA dataset with AI-generated reasoning paths, combined with hybrid-reward reinforcement learning, improves a 7B vision-language model across 14 benchmarks.

  9. Enhancing Spatial Reasoning in Vision-Language Models via Chain-of-Thought Prompting and Reinforcement Learning

    cs.CV 2025-07 conditional novelty 4.0 of 10

    Scene-graph-based chain-of-thought prompting and GRPO training improve spatial reasoning accuracy in vision-language models, and GRPO degrades less than supervised fine-tuning when question wording is flipped.

  10. Reinforcement Learning Meets Large Language Models: A Survey of Advancements and Applications Across the LLM Lifecycle

    cs.CL 2025-09 conditional novelty 3.0 of 10

    A survey that maps reinforcement learning methods, datasets, benchmarks, and open-source tools across the full training lifecycle of large language models, focusing on verifiable-reward reasoning.

Pith tools