# ReasonCLIP-58M: Visually Grounded Commonsense Reasoning Supervision for CLIP

arXiv: https://arxiv.org/abs/2606.26794 · HTML: https://mlsanity.com/paper/2606.26794/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 34 of its 117 references that are in the map.
- First posted on arXiv 25 Jun 2026.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)

## Builds on (most cited first, 30 of 34)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models](https://mlsanity.com/paper/2306.13394/) (2023)
- [LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs](https://mlsanity.com/paper/2111.02114/) (2021)
- [DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection](https://mlsanity.com/paper/2203.03605/) (2022)
- [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https://mlsanity.com/paper/2502.14786/) (2025)
- [OpenAI GPT-5 System Card](https://mlsanity.com/paper/2601.03267/) (2025)
- [EVA-CLIP: Improved Training Techniques for CLIP at Scale](https://mlsanity.com/paper/2303.15389/) (2023)
- [When and why vision-language models behave like bags-of-words, and what to do about it?](https://mlsanity.com/paper/2210.01936/) (2022)
- [LLaVA-CoT: Let Vision Language Models Reason Step-by-Step](https://mlsanity.com/paper/2411.10440/) (2024)
- [Supervision Exists Everywhere: A Data Efficient Contrastive Language-Image Pre-training Paradigm](https://mlsanity.com/paper/2110.05208/) (2021)
- [WIT: Wikipedia-based Image Text Dataset for Multimodal Multilingual Machine Learning](https://mlsanity.com/paper/2103.01913/) (2021)
- [Demystifying CLIP Data](https://mlsanity.com/paper/2309.16671/) (2023)
- [Data Filtering Networks](https://mlsanity.com/paper/2309.17425/) (2023)
- [Video-MMMU: Evaluating Knowledge Acquisition from Multi-Discipline Professional Videos](https://mlsanity.com/paper/2501.13826/) (2025)
- [E5-V: Universal Embeddings with Multimodal Large Language Models](https://mlsanity.com/paper/2407.12580/) (2024)
- [Scalable Pre-training of Large Autoregressive Image Models](https://mlsanity.com/paper/2401.08541/) (2024)
- [VisuLogic: A Benchmark for Evaluating Visual Reasoning in Multi-modal Large Language Models](https://mlsanity.com/paper/2504.15279/) (2025)
- [EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters](https://mlsanity.com/paper/2402.04252/) (2024)
- [What If We Recaption Billions of Web Images with LLaMA-3?](https://mlsanity.com/paper/2406.08478/) (2024)
- [FG-CLIP: Fine-Grained Visual and Textual Alignment](https://mlsanity.com/paper/2505.05071/) (2025)
- [Meta CLIP 2: A Worldwide Scaling Recipe](https://mlsanity.com/paper/2507.22062/) (2025)
- [Jina CLIP: Your CLIP Model Is Also Your Text Retriever](https://mlsanity.com/paper/2405.20204/) (2024)
- [jina-clip-v2: Multilingual Multimodal Embeddings for Text and Images](https://mlsanity.com/paper/2412.08802/) (2024)
- [CLIPS: An Enhanced CLIP Framework for Learning with Synthetic Captions](https://mlsanity.com/paper/2411.16828/) (2024)
- [Preserving Multi-Modal Capabilities of Pre-trained VLMs for Improving Vision-Linguistic Compositionality](https://mlsanity.com/paper/2410.05210/) (2024)
- [LLM2CLIP: Powerful Language Model Unlocks Richer Cross-Modality Representation](https://mlsanity.com/paper/2411.04997/) (2024)

## Abstract (arXiv)

CLIP and its variants are widely adopted visual backbones in multimodal systems, but their pretraining remains dominated by descriptive image-text alignment. As downstream applications increasingly demand visually grounded commonsense inference and compositional reasoning, it remains unclear whether CLIP-style encoders can support such reasoning without architectural changes. To address this, we present ReasonCLIP-58M, a continual pretraining framework that integrates large-scale reasoning supervision into CLIP-style models through our two-stage strategy, which progressively integrates reasoning signals while preserving descriptive alignment, followed by category-structured reasoning supervision. To support this framework, we construct two complementary datasets and a benchmark: ReasonLite-42M, with open-form, visually verifiable reasoning captions; ReasonPro-16M, with category-specific reasoning supervision; and RCLIP-Bench for diagnostic evaluation of visually grounded reasoning. We train a family of ReasonCLIP that improves visually grounded commonsense and compositional reasoning while also enhancing zero-shot retrieval performance. As a drop-in visual encoder for multimodal large language models such as LLaVA-NeXT, ReasonCLIP delivers consistent gains without additional inference cost, demonstrating that structured reasoning supervision enhances the expressive capacity of CLIP-style visual representations. All datasets, models, and training code are available at https://github.com/RISys-Lab/ReasonCLIP.