LVLM

2 items

RESEARCHDEV.to AI·4/22/2026

InternLM-XComposer-2.5: A Versatile Large Vision Language Model SupportingLong-Contextual Input and Output

InternLM-XComposer-2.5 is introduced as a versatile Large Vision Language Model. Its notable capability is supporting long-contextual input and output.

AI models multimodal AI LVLM Large Vision Language Model

RESEARCHarXiv CS.CL·4/15/2026

Benchmarking Deflection and Hallucination in Large Vision-Language Models

This paper introduces VLM-DeflectionBench, a new benchmark for Large Vision-Language Models (LVLMs) focusing on deflection and hallucination when dealing with conflicting or insufficient evidence. It also proposes a dynamic data curation pipeline to maintain benchmark difficulty over time and a fine-grained evaluation protocol to disentangle model behavior.

Retrieval Augmented Generation hallucination Benchmarking LVLM