AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

CrossHOI-Bench: A Unified Benchmark for HOI Evaluation across Vision-Language Models and HOI-Specific Methods

Qinqian Lei, Bo Wang, Robby T. Tan

CVPR 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Open-vocabulary alignment from VLMs · 1 citations (OpenAlex)

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 changedWhereMeasuredEffect
evaluate with the target person's box given instead of requiring detection (Setting 2 vs 1)Qwen2.5-VL-7B · CrossHOI-BenchInstance-F1: 30.53 -> 57.25Improved
evaluate with the target person's box given instead of requiring detection (Setting 2 vs 1)InternVL3-8B · CrossHOI-BenchInstance-F1: 4.96 -> 52.35Improved
evaluate with the target person's box given instead of requiring detection (Setting 2 vs 1)Qwen3-VL-30B-Instruct · CrossHOI-BenchInstance-F1: 0.25 -> 64.00Improved
evaluate with the target person's box given instead of requiring detection (Setting 2 vs 1)Qwen2.5-VL-32B · CrossHOI-BenchInstance-F1: 52.94 -> 69.52Improved
require recognizing interactions of all people (Setting 3 vs 2) instead of the target person onlyVLMs · CrossHOI-BenchEM: decreases by 5-10Hurt
require recognizing interactions of all people (Setting 3 vs 2) instead of the target person onlyInternVL3-8B · CrossHOI-Benchaverage precision: decreases by 9.1%Mixed
remove overly simple head-class scenes from HICO-DET test setCrossHOI-Bench test set · HICO-DET train vs testKL divergence: 0.088 -> 0.629Improved
use top-5 selection for HOI-specific models rather than threshold filtering (ablation)HOI-specific methods · CrossHOI-Benchmatching strategyImproved
compare with HOI-specific methods agreement on coarse-screened negativesHOI-specific methods vs VLMs · CrossHOI-Bench negativesagreement: ~99% vs. 95-97%Improved
use the CrossHOI-Bench subset rather than the full HICO-DET test setVLMs and HOI-specific methods · HICO-DET-based evaluationperformanceHurt
fine-tune Qwen2.5-VL-7B on HICO-DET-based train set (SFT)Qwen2.5-VL-7B · V-COCO and SWiG-HOI sub-benchmarksgeneralizationImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy