Papers

Filtered to open-source architectures · clear filter

Browse by term

continual learning 64reinforcement learning 34large language models 17benchmarking 12vision-language models 10generative models 8language models 8video generation 7multimodal models 6natural language processing 6robotics 6world models 6benchmarks 5diffusion models 5on-policy distillation 5policy optimization 5scalability 5self-distillation 5vision-language-action models 5autoregressive models 4computer vision 4diffusion transformers 4LLMs 4multimodal large language models 4verifiable rewards 4attention mechanisms 3embodied intelligence 3image editing 3long-term memory 3multimodal learning 3

Matching papers

MODUS: Decoder-Only Any-to-Any Modeling of Diverse Modalities

13 upvotes · 28 JUL 2026 · Mingqiao Ye, Zhaochong An, Zhitong Gao et al.

This paper introduces Modus, a new type of any-to-any model that can predict any modality from any combination of others using only a decoder, without the need for specialized heads or losses. This approach can support a wide range of applications, such as chained generation and cross-modal self-verification, and has shown strong out-of-the-box performance on various benchmarks.

multimodal modelsdecoder-only modelsany-to-any modelingvision-language modelsopen-source architectures