arXiv:2506. 03922v4 Announce Type: replace-cross Abstract: Multimodal Large Language Models (MLLMs) have demonstrated significant potential to advance a broad range of domains.
arXiv:2511. 17731v2 Announce Type: replace-cross Abstract: Chain-of-Thought (CoT) prompting has proven remarkably effective for eliciting complex reasoning in large language models (LLMs).
arXiv:2507. 19634v4 Announce Type: replace-cross Abstract: Recent advances in large language models have laid the foundation for multimodal LLMs (MLLMs), which unify text, speech, and vision within a single framework.
BEAR-Bench is a bilingual benchmark for multimodal large language models, featuring 1,000 human‑annotated questions derived from text‑rich business and scientific documents in English and Russian. It evaluates 16 MLLMs, including Gemini 3.1 Pro and Qwen3.5‑397B, revealing significant performance gaps even for the strongest systems. The benchmark also serves to compare hallucination‑detection methods by analyzing model failures on these complex documents.
SciMIF is a new benchmark that evaluates how well multimodal large language models (MLLMs) can follow complex scientific instructions. It is built on an analysis of 22 tasks across five scientific fields and introduces a taxonomy of 10 constraint groups that capture both general and discipline‑specific requirements. Experiments show large performance gaps between fields—chemistry is hardest—and that larger models do not necessarily improve constraint adherence, especially for fine‑grained, knowledge‑heavy instructions.
DocHop is a new benchmark that tests multimodal large language models on integrated chart‑context reasoning within document‑style images. The benchmark presents narrative text that imposes multi‑step compositional constraints, while charts supply the data needed to answer questions grounded in semantic reference labels. It contains 2,074 examples across six task categories, generated via a stochastic logic‑first pipeline that controls reasoning depth and visual density, and shows a large performance gap between humans (over 90% accuracy) and the best models (62.83%).
Lingshu is a medical‑specialized multimodal large language model that addresses key limitations of existing medical MLLMs, such as narrow knowledge coverage, hallucinations, and weak reasoning. The authors curate a comprehensive dataset combining medical imaging, texts, and general‑domain data, then train Lingshu in multiple stages to embed medical expertise and improve task performance. They also introduce MedEvalKit, a unified evaluation framework, and demonstrate that Lingshu outperforms current open‑source multimodal models on multimodal QA, text‑based QA, and medical report generation.
arXiv:2511. 18121v2 Announce Type: replace-cross Abstract: While Multimodal Large Language Models (MLLMs) excel on benchmarks, their processing paradigm differs from the human ability to integrate visual information.
arXiv:2604. 18347v2 Announce Type: replace-cross Abstract: Vision Language Models (VLMs) achieved rapid progress in the recent years.
arXiv:2507. 16518v3 Announce Type: replace-cross Abstract: Recent advances in multimodal large language models (MLLMs) have shown impressive reasoning capabilities.
arXiv:2606. 08034v1 Announce Type: cross Abstract: Symbolic benchmarks have emerged as a key approach to assess model robustness under minor modifications to STEM-related questions.
arXiv:2606. 12809v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) are trained on massive multimodal data, making data unlearning increasingly important as data owners may request the removal of specific content.
CoVA‑SFT is a new large‑scale dataset comprising 51.9K samples and over 222K multimodal reasoning steps that teach models to interleave text and visual abstractions across five layout families and 17 complex tasks. It includes explicit rationale formulations, agentic renderings, and verification loops to help models build and maintain internal visual workspaces for purely textual reasoning problems. A companion benchmark, CoVA‑Bench, contains 1,700 held‑out test samples for reproducible evaluation, and models fine‑tuned on CoVA‑SFT outperform all interleaved CoT baselines by more than 2× on average, though they still lag behind strong text‑only CoT baselines.