MLLMs
PulseAugur coverage of MLLMs — every cluster mentioning MLLMs across labs, papers, and developer communities, ranked by signal.
- instance of Multimodal LLMs 95%
- instance of alphaXiv 90%
- instance of Gotit.pub 90%
- instance of CatalyzeX 90%
- instance of Multimodal Large Language Models and Tunings: Vision, Language, Sensors, Audio, and Beyond 90%
- instance of DagsHub 90%
- instance of Qwen2.5-VL 90%
- instance of multimodal large language model 90%
- instance of ScienceCast 70%
- developed alphaXiv 70%
- developed Multimodal Large Language Models and Tunings: Vision, Language, Sensors, Audio, and Beyond 70%
- used by Gotit.pub 70%
- 2026-05-22 research_milestone A new pipeline was introduced to enhance MLLMs for safety-critical driving video analysis. source
- 2026-05-22 research_milestone Researchers reveal and propose a method to recover temporal grounding in multimodal large language models. source
- 2026-05-22 research_milestone A new benchmark and dataset were introduced to evaluate MLLMs' ability to reason about personality beyond superficial cues. source
- 2026-05-21 research_milestone A new method using MLLMs for detecting AI-generated Chinese poetry achieves state-of-the-art results. source
22 day(s) with sentiment data
-
New framework uses LLMs for improved diagnosis of cystic lung diseases
Researchers have developed DCLDs-EIR, a novel framework designed to improve the diagnosis of Diffuse Cystic Lung Diseases (DCLDs). This system leverages Multimodal Large Language Models (MLLMs) to integrate imaging and …
-
MLLMs tested as human surrogates for Explainable AI evaluation
Researchers have explored the use of Multimodal Large Language Models (MLLMs) as a substitute for human evaluation in the field of Explainable AI (XAI). A new framework was developed to measure explanation utility for b…
-
New OpenView system boosts MLLMs' out-of-view image understanding
Researchers have introduced OpenView, a novel pipeline designed to enhance multimodal large language models (MLLMs) in understanding content beyond the visible frame of an image, a capability known as out-of-view (OOV) …
-
New benchmark evaluates AI's ability to generate animated webpages from video
Researchers have introduced AnimWeb-Bench, a new benchmark designed to evaluate the capabilities of multimodal large language models (MLLMs) and coding agents in generating animated webpages from video input. This bench…
-
V-STORM dataset enhances MLLM spatio-temporal video reasoning
Researchers have introduced V-STORM, a novel framework designed to enhance spatio-temporal reasoning in Multimodal Large Language Models (MLLMs). This framework converts videos into Spatio-Temporal Scene Graphs (STSGs) …
-
MammoSeek agent mimics radiologists with selective zooming for mammography reports
Researchers have developed MammoSeek, a novel mammography agent designed to mimic the diagnostic process of radiologists. Unlike existing MLLMs that view images at a fixed resolution, MammoSeek employs a selective zoomi…
-
New PhysTrace framework evaluates physical consistency in video generation models
Researchers have introduced PhysTrace, a novel framework designed to evaluate the physical consistency of video generative models. This system utilizes Multimodal Large Language Models (MLLMs) to establish history-condi…
-
New PSGAgent framework enhances MLLMs' visual perception for scene graph generation
Researchers have introduced PSGAgent, a novel framework designed to improve the visual perception capabilities of Multimodal Large Language Models (MLLMs) for Panoptic Scene Graph generation. This agentic approach utili…
-
New ProQuant framework enables efficient low-bit quantization for edge MLLMs
Researchers have developed ProQuant, a new framework for quantization-aware training designed to make multi-modal large language models (MLLMs) more efficient for edge devices. This method addresses the performance degr…
-
New framework tackles MLLM hallucinations using entropy dynamics
Researchers have developed a novel framework to address hallucinations in multimodal large language models (MLLMs) by analyzing the dynamics of information entropy during the reasoning process. The study identifies spec…
-
New BEAKER benchmark tests AI's embodied reasoning in chemical labs
Researchers have introduced BEAKER, a new benchmark designed to evaluate the embodied capabilities of Multimodal Large Language Models (MLLMs) in self-driving chemical laboratories. The benchmark consists of 1,000 exper…
-
New benchmark and debate framework improve scam video detection
Researchers have developed MechScam, a new benchmark dataset with over 6,600 videos designed to improve the detection of online scam videos. Unlike previous methods that focused on simple classification, MechScam provid…
-
New BEAKER benchmark tests AI's embodied reasoning in chemical labs
Researchers have introduced BEAKER, a new benchmark designed to evaluate the embodied capabilities of Multimodal Large Language Models (MLLMs) in self-driving chemical laboratories (SDCLs). BEAKER comprises 1,000 expert…
-
New DSRS-Net architecture teaches MLLMs reasoning for driving scenes
Researchers have introduced DSRS-Net, a novel architecture designed for driving scene reasoning segmentation (DSRS). This new approach aims to equip multimodal large language models (MLLMs) with scene-specific awareness…
-
New DGP method enhances MLLMs by connecting all vision and language model layers
Researchers have developed Dynamic Gated Pathways (DGP), a novel routing mechanism for Multimodal Large Language Models (MLLMs). DGP establishes all-to-all connections between every layer of a vision encoder and a langu…
-
MLLMs for Video Analysis: Input Representation Matters for Social Science Research
A new study published on Hugging Face explores the effectiveness of multimodal large language models (MLLMs) for social science research involving video analysis. The research, which tested six open-weight MLLMs across …
-
New VideoMemBench benchmark reveals MLLMs struggle with in-context video memory
A new benchmark called VideoMemBench has been developed to evaluate the in-context video memory of multimodal large language models (MLLMs). This benchmark assesses the models' ability to recall information from earlier…
-
FactEmbed enhances Composed Image Retrieval using factorized MLLM embeddings · 4 sources tracked
Researchers have introduced FactEmbed, a novel approach to Composed Image Retrieval (CIR) that enhances the utilization of Multimodal Large Language Models (MLLMs). Unlike previous methods that directly map composed que…
-
Open-weight MLLMs struggle with visual comparison tasks, paper finds
A new arXiv paper investigates the limitations of open-weight Multimodal Large Language Models (MLLMs) in performing visual comparisons, particularly in biological image analysis. Researchers found that even simple task…
-
New ReVEAL framework uses MLLMs for reasoning video object segmentation
Researchers have developed ReVEAL, a novel framework for reasoning video object segmentation (ReVOS) that decouples the MLLM reasoning process from the segmentation model. This approach uses Visual Evidence Tokens (VETs…