dMLLM-TTS delivers up to 6x more efficient test-time scaling for diffusion MLLMs via O(N+T) hierarchical search and self-verified feedback, improving generation quality on GenEval across three models.
amused: An open muse reproduction
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
representative citing papers
A new formulation for composing discrete generative processes enables precise control over novel condition combinations in image generation, cutting error rates by 63% and speeding up inference.
A derived formulation for composing discrete probabilistic generative processes enables novel condition combinations in image generation, yielding 63.4% relative error reduction and FID gains on CLEVR and FFHQ datasets.
Presents open-source 7B models for million-token video and language understanding via Blockwise RingAttention, setting new benchmarks in retrieval and long video tasks.
citing papers explorer
-
dMLLM-TTS: Self-Verified and Efficient Test-Time Scaling for Diffusion Multi-Modal Large Language Models
dMLLM-TTS delivers up to 6x more efficient test-time scaling for diffusion MLLMs via O(N+T) hierarchical search and self-verified feedback, improving generation quality on GenEval across three models.
-
Controllable Image Generation with Composed Parallel Token Prediction
A new formulation for composing discrete generative processes enables precise control over novel condition combinations in image generation, cutting error rates by 63% and speeding up inference.
-
Controllable Image Generation with Composed Parallel Token Prediction
A derived formulation for composing discrete probabilistic generative processes enables novel condition combinations in image generation, yielding 63.4% relative error reduction and FID gains on CLEVR and FFHQ datasets.
-
World Model on Million-Length Video And Language With Blockwise RingAttention
Presents open-source 7B models for million-token video and language understanding via Blockwise RingAttention, setting new benchmarks in retrieval and long video tasks.