ECCV 2026 accepted papers on Large language models
31 of the 2,834 papers accepted at ECCV 2026 (1.1%) are on Large language models, compared with 0.5% at ECCV 2024. They fall into 7 topics.
See ECCV 2026 in the interactive map
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 · 10 papers
- MGM-Omni: Scaling Omni LLMs to Personalized Long-Horizon Speech (10 citations)
- PuzLM: Solving Jigsaw Puzzles with Sequence-to-Sequence Language Models (6 citations)
- SignBind-LLM: Multi-Stage Modality Fusion for Sign Language Translation (4 citations)
- SIGNER: Temporally Grounded Sign Language Generation via Time-Resolved Conditioning (4 citations)
- BackTranslation2.0 - A Linguistically Motivated Metric to Assess Sign Language Production (2 citations)
- Pointer-CAD v2: Plan-Then-Construct CAD Generation with Dimension-Aware Parametric Precision (2 citations)
- SIGNET: Motion-Level Knowledge Transfer for Cross-Language Sign Language Translation (1 citation)
- Gaze-to-text Generation: Beyond Categorical Decoding of Human Attention
- Q-BridgeNet: A Quantization Network for Cross-Lingual Sign Language Translation
- Asymmetric Anchoring: Opening the Black Box of MLLMs for Forgery Detection
Reasoning and chain of thought · 9 papers
- SMART: When is it Actually Worth Expanding a Speculative Tree? (3 citations)
- CritiqueDriveVLM: From Verifier-Guided Reinforcement Learning to Latent Thought Distillation for Autonomous Driving (2 citations)
- AnE: Pushing the Reasoning Frontier of Multimodal LLMs via Anchor Evolution (1 citation)
- CARE: Causally-Aligned Reasoning Exploration for Medical Large Language Models
- Learning Consistency in Reward Modeling for Multi-Modal Reasoning
- AdaThinking-E: One-Token Entropy Regulation for Adaptive Thinking
- Practice Makes Perfect: From Explicit Decomposition to Reinforced Latent Planning in Text-to-Human Motion
- Rethinking Training and Inference for Trajectory Forecasting: Linking Winner-Take-All back to GMMs
- Wan-R1: Verifiable-Reinforcement Learning for Generalizable Video Reasoning
