ALMs unify pretrained atomistic encoder, LLM, and denoising diffusion via continuous projectors and staged training to reach SOTA on text-conditioned crystal prediction and de novo generation.
arXiv:2406.07209
14 Pith papers cite this work. Polarity classification is still indexing.
14
Pith papers citing it
representative citing papers
Introduces PexelsCustom-1M dataset, CustoMDiT parameter-efficient model, and OpenCustom benchmark for open-domain customized video generation.
EM-Vid introduces an entity-centric latent patch memory bank with sparse token conditioning and budgeted updates for training-free consistent multi-shot video generation.
ASTRA disentangles subject identity from pose structure in diffusion transformers via retrieval-augmented pose guidance, asymmetric EURoPE embeddings, and a DSM adapter to improve multi-subject generation.
LooseRoPE modulates RoPE in diffusion attention maps to continuously trade off between preserving a pasted object's identity and harmonizing it with its new surroundings.
T2I-FactualBench is a new three-tier benchmark for factuality of knowledge-intensive concepts in T2I models, using multi-round VQA evaluation to show SOTA models need improvement.
MIBE introduces a multi-subject interaction benchmark (MIB) with silver and gold sets and a dual-head evaluator (MIE) trained on VLM labels that outperforms baselines in matching human judgments.
Presents MRT, a 20B-parameter masked region diffusion model unifying text-to-layers, image-to-layers, and layers-to-layers tasks with an overflow-aware canvas layer for complete editable outputs.
A zero-shot subject-driven video generation framework that decomposes the task into identity injection from 200K subject-image pairs and motion preservation from 4K arbitrary videos, trained in 288 A100 GPU hours on CogVideoX-5B to match prior performance at 1% compute.
FreeGraftor performs subject-driven text-to-image generation without training by cross-image feature grafting via semantic matching, position-constrained attention fusion, and a noise initialization strategy that preserves reference geometry.
A data-generation pipeline plus pairwise subject-consistency rewards in RL improve consistency and prompt adherence for multi-subject personalized image generation.
AnimalBooth introduces an Animal Net, adaptive attention module, and frequency-controlled DCT feature integration to improve identity preservation and perceptual quality in personalized animal image generation, supported by a new high-resolution dataset AnimalBench.
UniVerse proposes a unified modulation framework for segmentation-free, disentangled multi-concept personalization in diffusion transformers, claiming superior localization and fidelity over baselines.
citing papers explorer
-
T2I-FactualBench: Benchmarking the Factuality of Text-to-Image Models with Knowledge-Intensive Concepts
T2I-FactualBench is a new three-tier benchmark for factuality of knowledge-intensive concepts in T2I models, using multi-round VQA evaluation to show SOTA models need improvement.