How I Study AI - Learn AI Papers & Lectures the Easy Way

Omni-R1: Towards the Unified Generative Paradigm for Multimodal Reasoning

Intermediate

Dongjie Cheng, Yongqi Li et al.Jan 14arXiv

Omni-R1 teaches AI to think with pictures and words at the same time by drawing helpful mini-images while reasoning.

#multimodal reasoning#interleaved generation#functional image generation

RelayLLM: Efficient Reasoning via Collaborative Decoding

Intermediate

Chengsong Huang, Tong Zheng et al.Jan 8arXiv

RelayLLM lets a small model do the talking and only asks a big model for help on a few, truly hard tokens.

#token-level collaboration#<call>n</call> command#collaborative decoding

Factorized Learning for Temporally Grounded Video-Language Models

Intermediate

Wenzheng Zeng, Difei Gao et al.Dec 30arXiv

This paper teaches video-language models to first find when the proof happens in a video and then answer with that proof, instead of mixing both steps together.

#temporal grounding#video-language models#evidence tokens

Papers3

Omni-R1: Towards the Unified Generative Paradigm for Multimodal Reasoning

RelayLLM: Efficient Reasoning via Collaborative Decoding

Factorized Learning for Temporally Grounded Video-Language Models