AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

NAVER: A Neuro-Symbolic Compositional Automaton for Visual Grounding with Explicit Logic Reasoning

Zhixi Cai, Fucai Ke, Simindokht Jahangard, Maria Garcia de la Banda, Reza Haffari, Peter J. Stuckey, Hamid Rezatofighi

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Chain-of-thought multimodal reasoning · 8 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
add DFA self-correction transitions (ablation)NAVER · RefCOCO testAaccuracy: 57.4 -> 78.1Improved
add DFA self-correction transitions (ablation)NAVER · Ref-Adv testaccuracy: 43.6 -> 75.4Improved
add logic generation and reasoning (ablation)NAVER · RefCOCO+ testAaccuracy: 51.3 -> 62.0Improved
add answerer validation (ablation)NAVER · RefCOCO testAaccuracy: 51.7 -> 57.4Improved
add answerer validation to DFA + logic (ablation)NAVER · RefCOCOg testaccuracy: 75.8 -> 91.6Improved
add captioner before entity category extraction (ablation)NAVER · RefCOCOg testaccuracy: 80.2 -> 91.6Improved
add captioner before entity category extraction (ablation)NAVER · RefCOCO testAaccuracy: 91.6 -> 96.2Improved
use NAVER with GLIP-L instead of GLIP-L aloneGLIP-L · RefCOCO testAaccuracy: 55.0 -> 70.1Improved
wrap Florence2-L in NAVERFlorence2-L · RefCOCO testAaccuracy: 95.1 -> 96.2Improved
use NAVER instead of HYDRA (same models)HYDRA · RefCOCOg testruntime failure rate (%): 10.3 -> 00.3Improved
swap VLM among BLIP2-XXL, xGen-MM, LLaVA1.5-7B, InternVL2-8B (ablation)NAVER · RefCOCO testAaccuracy: 94.5 -> 96.2No clear effect
swap LLM from GPT-4o Mini to DeepSeek-R1 70B (ablation)NAVER · RefCOCOg detectionaccuracy: 91.6 -> 85.8Hurt

Later papers that compared with this method

1 later accepted paper reports NAVER as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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