Research map / Vision-language and multimodal / Multimodal LLMs
Strengthening Multimodal Large Language Model with Bootstrapped Preference Optimization
Renjie Pi, Tianyang Han, Wei Xiong, Jipeng ZHANG, Runtao Liu, Rui Pan, Tong Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| apply BPO preference optimization to LLaVA1.5-7B | LLaVA1.5-7B · MM-Vet | Total score: 31.7 -> 36.8 | Improved |
| apply BPO to LLaVA1.5-13B | LLaVA1.5-13B · MM-Vet | Total score: 36.8 -> 41.4 | Improved |
| apply BPO to reduce hallucination | LLaVA1.5-7B · Object-HalBench | Response-level hallucination rate (%): 45.9 -> 31.9 | Improved |
| apply BPO to reduce object-level hallucination | LLaVA1.5-13B · Object-HalBench | Object-level hallucination rate (%): 21.8 -> 12.9 | Improved |
| use preference learning with negatives instead of SFT on positive responses only (ablation) | LLaVA1.5-7B · MM-Vet | Total score: 33.3 -> 36.8 | Improved |
| use image-weakened negatives instead of self-generated negatives without weakening (ablation) | LLaVA1.5-7B BPO · MM-Vet | Total score: 32.4 -> 34.3 | Improved |
| increase preference dataset size (ablation) | BPO · MM-Vet (LLaVA-7B, LLaVA-13B) | Total score | Improved |
| BPO versus SFT at all dataset sizes (ablation) | LLaVA-7B and 13B · MM-Vet | Total score | Improved |
| self-generated negatives (no weakening) on math capability (ablation) | LLaVA1.5-7B · MM-Vet Math | Math score: 7.70 -> 1.90 | Hurt |
Closest papers by meaning
- Debiasing Multimodal Large Language Models via Noise-Aware Preference Optimization (CVPR 2025 · similarity 0.90)
- Task Preference Optimization: Improving Multimodal Large Language Models with Vision Task Alignment (CVPR 2025 · similarity 0.88)
- Multimodal Prompt Optimization: Why Not Leverage Multiple Modalities for MLLMs (ICLR 2026 · similarity 0.88)
- Language Bias in LVLMs: From In-Depth Analysis to Simple and Effective Mitigation (ICML 2026 · similarity 0.88)
- Zooming from Context to Cue: Hierarchical Preference Optimization for Multi-Image MLLMs (NeurIPS 2025 · similarity 0.88)
- CHiP: Cross-modal Hierarchical Direct Preference Optimization for Multimodal LLMs (ICLR 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
