MMEB-V2
PulseAugur coverage of MMEB-V2 — every cluster mentioning MMEB-V2 across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
New PILOT framework enhances multimodal embeddings via guided reasoning exploration
Researchers have developed PILOT, a new framework designed to enhance multimodal embeddings by improving the reasoning process used in their generation. Unlike previous methods that relied on post-hoc rewards, PILOT mod…
-
System One decision models show promise in legal, clinical, and multimodal AI tasks
Researchers are exploring the application of "System One" decision models, which are derived from large language models but focus on outputting calibrated probabilities rather than generating text, for specialized tasks…
-
VaME framework introduces variational latent reasoning for multimodal embeddings
Researchers have introduced VaME, a novel framework for multimodal embeddings that utilizes variational latent reasoning. Unlike deterministic approaches, VaME models latent reasoning as a learnable distribution over tr…
-
New frameworks enhance LVLM reasoning with improved credit assignment and efficiency
Two new research papers propose novel frameworks for enhancing the reasoning capabilities of large vision-language models (LVLMs). The first paper, PIVOT, introduces a dual-level learning framework that uses self-calibr…
-
MoEMB scales multimodal embeddings with efficient Mixture-of-Experts models
Researchers have introduced MoEMB, a novel approach to scaling universal multimodal embeddings using an efficient mixture-of-experts (MoE) architecture. This method allows for increased encoder capacity while maintainin…
-
New MMEmb-R1 framework enhances multimodal embedding with adaptive reasoning
Researchers have introduced MMEmb-R1, a novel framework designed to enhance multimodal embedding by adaptively incorporating reasoning capabilities. This approach addresses challenges in MLLMs by selectively applying ch…
-
Tencent unveils WeMM-Embedding multimodal models for WeChat
Tencent has introduced WeMM-Embedding, a new family of universal multimodal embedding models designed to represent diverse content like text, images, and videos in a shared space. Available in 2B, 4B, and 9B variants, t…
-
WeChat releases WeMM-Embedding multimodal models, achieving SOTA performance
WeChat has introduced WeMM-Embedding, a new family of universal multimodal embedding models designed to represent text, images, videos, and interleaved multimodal inputs in a shared space. The models, available in 2B, 4…
-
UMER framework unifies embedding and ranking for multimodal retrieval
Researchers have introduced UMER, a novel framework designed to unify embedding and ranking for universal multimodal retrieval tasks. This approach employs Pair-Aware Discriminative Reasoning, which contrasts query-cand…
-
New UniME-R1 framework improves multimodal retrieval with feedback-driven reasoning · 2 sources tracked
Researchers have developed UniME-R1, a novel framework designed to enhance unified multimodal retrieval by incorporating retrieval feedback into the reasoning process. Unlike previous methods that relied solely on query…
-
Douyin unveils advanced multimodal embedding model for search and recommendation · 2 sources tracked
Researchers have developed the Douyin Multimodal Embedding (DME) model, a two-stage system designed for efficient and fine-grained multimodal search and recommendation. The model first undergoes large-scale contrastive …
-
Douyin unveils new multimodal embedding model for enhanced search and recommendation
Researchers have introduced Douyin Multimodal Embedding (DME), a novel two-stage framework designed to enhance multimodal representation learning for large-scale platforms like Douyin, Xiaohongshu, and YouTube. The firs…
-
New models unify sparse and dense multimodal embeddings, boosting search efficiency
Researchers have introduced UEmbed, a novel decoder-only multimodal embedding model capable of generating both sparse lexical and dense representations within a single causal forward pass. This model aims to unify spars…
-
Eddy-VL 1.9B: Compressed multimodal model for edge deployment
Researchers have developed Eddy-VL 1.9B, a compressed multimodal embedding model designed for edge deployment in environments without cloud access. Built upon Qwen3-VL-Embedding-2B, Eddy-VL utilizes structural pruning a…
-
New SMART Framework Enhances Multimodal Retrieval with Latent Multi-Vector Capabilities
Researchers have introduced SMART, a framework designed to enhance multimodal retrieval by unlocking the hidden multi-vector capabilities within standard single-vector embedding models. This approach uses contrastive tr…
-
FreeRet framework turns multimodal LLMs into training-free retrievers
Researchers have developed FreeRet, a novel framework that enables multimodal large language models (MLLMs) to function as effective retrievers without requiring additional training. This plug-and-play system extracts s…
-
New RIME framework enhances multimodal embeddings by optimizing generation and retrieval.
Researchers have introduced Rewrite-driven Multimodal Embedding (RIME), a new framework designed to enhance generative multimodal embeddings. RIME addresses limitations in Chain-of-Thought reasoning by optimizing genera…