AI Paper Atlas

Research map / Benchmarks / LLaVA-Bench (In-the-Wild)

LLaVA-Bench (In-the-Wild) leaderboard

13 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on LLaVA-Bench (In-the-Wild). Best score: 89.8 (LLaVA-1.5-13B + RoVRM (best-of-n), AAAI 2025).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

LLaVA-Bench (In-the-Wild) · score

13 papers, each with its best reported number for its own method; higher is better.

#score ↑MethodPaperTopic
189.8LLaVA-1.5-13B + RoVRM (best-of-n)
LLaVA-1.5-13B
RoVRM: A Robust Visual Reward Model Optimized via Auxiliary Textual Preference Data
AAAI 2025 · Table 1
Multimodal LLMs
287.1Meteor-7B
InternLM2-7B
Meteor: Mamba-based Traversal of Rationale for Large Language and Vision Models
NeurIPS 2024 · Table 1
Multimodal LLMs
376.9SPHINX-2k
LLaMA-2-13B
"SPHINX: A Mixer of Weights, Visual Embeddings and Image Scales for Multi-modal Large Language Models"
ECCV 2024 · Table 1
Multimodal LLMs
476.24Self-Aug
LLaVA-1.5-13B
Self-Augmented Visual Contrastive Decoding
ICLR 2026 · Table 2
Multimodal LLMs
575.7Honeybee C-Abstractor (13B, M=256)
Vicuna-13B
Honeybee: Locality-enhanced Projector for Multimodal LLM
CVPR 2024 · Table: Comparison with other s
Multimodal LLMs
673VILA-13B
Llama-2-13B
VILA: On Pre-training for Visual Language Models
CVPR 2024 · Table 5
CLIP and zero-shot
772.6ShareGPT4V-7B
Vicuna-7B
ShareGPT4V: Improving Large Multi-Modal Models with Better Captions
ECCV 2024 · Table 3
Multimodal LLMs
868.1VAR
LLaVA-1.5-7B
See What You Are Told: Visual Attention Sink in Large Multimodal Models
ICLR 2025 · General vision-language result
Multimodal LLMs
967.4SENA (θ3)
LLaVA-1.5-7B
Beyond Human Data: Aligning Multimodal Large Language Models by Iterative Self-Evolution
AAAI 2025 · Table 3
Multimodal LLMs
1066.9SRF-LLaVA-1.5 7B
Vicuna-1.5-7B
Towards Self-Refinement of Vision-Language Models with Triangular Consistency
NeurIPS 2025 · Table 1
Multimodal LLMs
3 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.