ICCV 2025 accepted papers on Large language models
18 of the 2,701 papers accepted at ICCV 2025 (0.7%) are on Large language models, compared with 0.5% at ICCV 2023. They fall into 7 topics.
See ICCV 2025 in the interactive map
Submitting to ICCV? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Language, safety and interpretability · 5 papers
- SmolDocling: An ultra-compact vision-language model for end-to-end multi-modal document conversion (60 citations)
- Signs as Tokens: A Retrieval-Enhanced Multilingual Sign Language Generator (27 citations)
- VALLR: Visual ASR Language Model for Lip Reading (17 citations)
- Zero-AVSR: Zero-Shot Audio-Visual Speech Recognition with LLMs by Learning Language-Agnostic Speech Representations (13 citations)
- Enhancing Numerical Prediction of MLLMs with Soft Labeling (2 citations)
Agents, code and math · 5 papers
- Pi-GPS: Enhancing Geometry Problem Solving by Unleashing the Power of Diagrammatic Information (14 citations)
- Aligning Constraint Generation with Design Intent in Parametric CAD (10 citations)
- AdaDrive: Self-Adaptive Slow-Fast System for Language-Grounded Autonomous Driving (6 citations)
- CATP-LLM: Empowering Large Language Models for Cost-Aware Tool Planning (6 citations)
- OD-RASE: Ontology-Driven Risk Assessment and Safety Enhancement for Autonomous Driving (1 citation)
