This paper proposes a multimodal graph-based approach for constructing knowledge graphs from visually rich documents to improve multimodal question answering.

The paper introduces MLLM-Microscope, a system that analyzes the internal structure of multimodal large language models (MLLMs), finding that modality fusion significantly impacts the linearity and di…

This paper presents the benefits of visual pretraining for foundation model intelligence, outperforming text-only pretraining on multiple backbones and benchmarks.

This paper discusses the current understanding of Large Language Models (LLMs), their capabilities, and their relationship to human cognition, with a focus on emerging capabilities and mechanistic imp…

This survey provides a comprehensive taxonomy and vulnerability-centric analysis of adversarial attacks targeting Multimodal Large Language Models (MLLMs), offering an explanatory framework for enhanc…

This paper analyzes speech-text interleaved language models and reveals that they go through an implicit transcription phase in which spoken words become decodable as text in intermediate layers.

The paper proposes a pipeline for building a travel-domain reasoning language model using an expert-designed knowledge graph, achieving an 82.4% exact match on a benchmark.

This paper proposes Efficient Chain-of-Modality Reasoning (ECoM Reasoning), a framework to improve reasoning ability in spoken language models (SLMs) for mathematical question answering tasks by compr…

The paper introduces FRANZ, a communicative audit framework, to evaluate how LLMs frame responses to subjective questions, finding that LLMs exhibit statistically significant and coupled differences i…

This paper introduces KGERMAR, a framework that constructs dynamic, context-specific knowledge graphs during inference for long-context language modeling, achieving lower perplexity and better memory…

This paper proposes a multi-turn retrieval-augmented generation pipeline for conversational systems across four domains.

This paper proposes a joint BERT-GNN architecture to systematically extract entities and relationships from diverse historical texts, achieving superior performance over conventional methods.

MemGraphRAG introduces a novel memory-based multi-agent system to construct globally consistent and structurally sound knowledge graphs, significantly improving retrieval-augmented generation for comp…

The paper builds a benchmark to evaluate the ability of multimodal large language models to extract accurate data tables from chart images, and proposes a human-centered approach to improve numerical…

This paper evaluates the performance of large language models on quiz-style questions covering common and niche topics in six European languages, finding significant knowledge gaps and language-depend…

This paper proposes BrainAgent, an agentic LLM framework for knowledge-enhanced brain network analysis, which improves performance and produces more comprehensive, multi-level, and verifiable explanat…

The paper introduces MechVQA, a comprehensive dataset and benchmark for mechanical drawing understanding, and proposes the MechVL model, which significantly improves Multimodal LLMs' performance on th…

The paper identifies a failure mode called spatial lexical bias in MLLMs, where adding a spatial word to options biases the model's choice, and demonstrates that this failure originates primarily from…

This paper proposes a lightweight encoder-based MEL solution called FAST-MEL that meets three objectives: high linking accuracy, computational efficiency, and storage efficiency.