Multimodal Reasoning

Video-R1: First to explore rule-based RL for video reasoning in MLLMs, enhancing performance on key benchmarks.

MLLMs still struggle with spatial reasoning! LEGO-Puzzles benchmark reveals critical deficiencies, paving the way for AI advancement.

OpenVLThinker: Iteratively refining vision-language models for complex reasoning, bridging the gap to R1-style capabilities.

MAPS solves multimodal scientific problems better by combining multiple agents and Socratic learning.

OThink-MR1 enhances MLLM reasoning via dynamic reinforcement learning, achieving remarkable cross-task generalization!

TVC mitigates visual forgetting in multimodal LLMs, enhancing reasoning by strategically re-introducing and compressing visual information.

MicroVQA: A new benchmark to test visual-question-answering in microscopy-based research.

A comprehensive survey of multimodal chain-of-thought (MCoT) reasoning, bridging the gap in existing literature and fostering innovation towards multimodal AGI.

MPBench: Multimodal benchmark to identify errors in reasoning processes.

ProJudge: MLLM judges’ benchmark for sci-reasoning & instruction-tuning data to boost performance!

R1-Omni: RLVR enhances multimodal emotion recognition, boosting reasoning and generalization.

R2-T2: Boost multimodal MoE performance by re-routing experts in test-time, no retraining needed!

MMIR: A new benchmark to assess and improve multimodal reasoning models’ ability to detect inconsistencies in real-world content.

video-SALMONN-01: An open-source audio-visual LLM enhances video understanding with a novel reasoning-intensive dataset and the pDPO method, achieving significant accuracy gains.

InfiR: Efficient, small AI models rival larger ones in reasoning, slashing costs and boosting privacy for wider AI use.

ThinkDiff empowers text-to-image diffusion models with multimodal reasoning by aligning vision-language models to an LLM decoder, achieving state-of-the-art results on in-context reasoning benchmarks.

GPT models’ multimodal reasoning abilities are tracked over time on challenging visual puzzles, revealing surprisingly steady improvement and cost trade-offs.

Virgo: A new multimodal slow-thinking system, significantly improves MLLM reasoning by fine-tuning with text-based long-form thought data, demonstrating comparable performance to commercial systems.

M-STAR: a novel self-evolving training framework significantly boosts multimodal reasoning in large models without human annotation, achieving state-of-the-art results.

AR-MCTS: a novel framework boosting multimodal large language model reasoning by actively retrieving key supporting evidence and using Monte Carlo Tree Search for improved path selection and verificat…