Developing autonomous home robots controlled by natural language has long been a pursuit of human. While advancements in large language models (LLMs) and embodied intelligence make this goal closer, several challenges persist: the lack of a unified benchmark for more complex robot tasks, limited evaluation methods and metrics, data incompatibility between LLMs and mobile manipulation trajectories. To address these issues, we introduce Embodied Mobile Manipulation in Open Environments (EMMOE), which requires agents to interpret user instructions and execute long-horizon everyday tasks in continuous space. EMMOE seamlessly integrates high-level and low-level embodied tasks into a unified framework, along with three new metrics for more diverse assessment. Additionally, we collect EMMOE-100, which features in various task attributes, detailed process annotations, re-plans after failures, and two sub-datasets for LLM training. Furthermore, we design HomieBot, a sophisticated agent system consists of LLM with Direct Preference Optimization (DPO), light weighted navigation and manipulation models, and multiple error detection mechanisms. Finally, we demonstrate HomieBot's performance and the evaluation of different models and policies.

The alignment of large language models (LLMs) with human preferences has recently emerged as a focal area of research [53, 62]. Prominent techniques such as Reinforcement Learning from Human Feedback (RLHF) [47] and Direct Preference Optimization (DPO) [50] have demonstrated substantial efficacy. However, these methods require the optimization of individual policies, posing challenges such as high consumption of training resources. Inference-time alignment [27, 45] provides an efficient alternative through direct adjustment of the model's output distribution, thus avoiding the need for resource-intensive retraining. Despite its advantages, this approach still requires policy-specific value functions, limiting its scalability across different models. Additionally, the inference-time latency remains high, presenting further challenges to its practical deployment. In this paper, we investigate an efficient and policy-agnostic preference optimization method. We begin by reconsidering the objective of aligning with humans [53, 65]. As illustrated in Figure 1(a), the alignment process operates at the sentence level, focusing on adjusting key components of the generated content, such as style or format, to better reflect human intentions or values.

The field of advanced text-to-image generation is witnessing the emergence of unified frameworks that integrate powerful text encoders, such as CLIP and T5, with Diffusion Transformer backbones. Although there have been efforts to control output images with additional conditions, like canny and depth map, a comprehensive framework for arbitrary text-image interleaved control is still lacking. This gap is especially evident when attempting to merge concepts or visual elements from multiple images in the generation process. To mitigate the gap, we conducted preliminary experiments showing that large multimodal models (LMMs) offer an effective shared representation space, where image and text can be well-aligned to serve as a condition for external diffusion models. Based on this discovery, we propose Dream Engine, an efficient and unified framework designed for arbitrary text-image interleaved control in image generation models. Building on powerful text-to-image models like SD3.5, we replace the original text-only encoders by incorporating versatile multimodal information encoders such as QwenVL. Our approach utilizes a two-stage training paradigm, consisting of joint text-image alignment and multimodal interleaved instruction tuning. Our experiments demonstrate that this training method is effective, achieving a 0.69 overall score on the GenEval benchmark, and matching the performance of state-of-the-art text-to-image models like SD3.5 and FLUX.

Human motion generation has advanced markedly with the advent of diffusion models. Most recent studies have concentrated on generating motion sequences based on text prompts, commonly referred to as text-to-motion generation. However, the bidirectional generation of motion and text, enabling tasks such as motion-to-text alongside text-to-motion, has been largely unexplored. This capability is essential for aligning diverse modalities and supports unconditional generation. In this paper, we introduce PackDiT, the first diffusion-based generative model capable of performing various tasks simultaneously, including motion generation, motion prediction, text generation, text-to-motion, motion-to-text, and joint motion-text generation. Our core innovation leverages mutual blocks to integrate multiple diffusion transformers (DiTs) across different modalities seamlessly. We train PackDiT on the HumanML3D dataset, achieving state-of-the-art text-to-motion performance with an FID score of 0.106, along with superior results in motion prediction and in-between tasks. Our experiments further demonstrate that diffusion models are effective for motion-to-text generation, achieving performance comparable to that of autoregressive models.

Aligning with personalized preferences, which vary significantly across cultural, educational, and political differences, poses a significant challenge due to the computational costs and data demands of traditional alignment methods. In response, this paper presents Personalized Alignment at Decoding-time (PAD), a novel framework designed to align LLM outputs with diverse personalized preferences during the inference phase, eliminating the need for additional training. By introducing a unique personalized reward modeling strategy, this framework decouples the text generation process from personalized preferences, facilitating the generation of generalizable token-level personalized rewards. The PAD algorithm leverages these rewards to guide the decoding process, dynamically tailoring the base model's predictions to personalized preferences. Extensive experimental results demonstrate that PAD not only outperforms existing training-based alignment methods in terms of aligning with diverse preferences but also shows significant generalizability to preferences unseen during training and scalability across different base models. This work advances the capability of LLMs to meet user needs in real-time applications, presenting a substantial step forward in personalized LLM alignment. Recent advancements have demonstrated success in aligning language models with human preferences and values (Stiennon et al., 2020; Bai et al., 2022; Ouyang et al., 2022; Achiam et al., 2023). However, in this pluralistic world, users' preferences can diverge significantly based on their different cultures, educational backgrounds, religions, and political stands (Gordon et al., 2022; Sorensen et al., 2024b; Jang et al., 2023; Cheng et al., 2023). Furthermore, even for the same person, the preference of a particular LLM response can vary when the application scenario changes. Hence, there always exists a proportion of human preferences that cannot be unified by the general preference, also known as personalized preferences, which current alignment frameworks struggle to align with due to the need for high-quality datasets and substantial computational costs in policy optimization. How can we align with personalized preferences without the need for additional data collection and policy training? In this paper, we introduce Personalized Alignment at Decoding-time (PAD), which aims to align LLM outputs with diverse personalized preferences during the inference phase without requiring additional training. To achieve this, we first propose a personalized reward modeling strategy, which decouples the text generation process (modeled as a Markov Decision Process) from personalized preferences, thereby enabling the acquisition of generalizable token-level personalized rewards.

Multimodal Large Language Model (MLLM) has recently garnered attention as a prominent research focus. By harnessing powerful LLM, it facilitates a transition of conversational generative AI from unimodal text to performing multimodal tasks. This boom begins to significantly impact medical field. However, general visual language model (VLM) lacks sophisticated comprehension for medical visual question answering (Med-VQA). Even models specifically tailored for medical domain tend to produce vague answers with weak visual relevance. In this paper, we propose a fine-grained adaptive VLM architecture for Chinese medical visual conversations through parameter-efficient tuning. Specifically, we devise a fusion module with fine-grained vision encoders to achieve enhancement for subtle medical visual semantics. Then we note data redundancy common to medical scenes is ignored in most prior works. In cases of a single text paired with multiple figures, we utilize weighted scoring with knowledge distillation to adaptively screen valid images mirroring text descriptions. For execution, we leverage a large-scale multimodal Chinese ultrasound dataset obtained from the hospital. We create instruction-following data based on text from professional doctors, which ensures effective tuning. With enhanced model and quality data, our Large Chinese Language and Vision Assistant for Ultrasound (LLaVA-Ultra) shows strong capability and robustness to medical scenarios. On three Med-VQA datasets, LLaVA-Ultra surpasses previous state-of-the-art models on various metrics.

With the power of large language models (LLMs), open-ended embodied agents can flexibly understand human instructions, generate interpretable guidance strategies, and output executable actions. Nowadays, Multi-modal Language Models~(MLMs) integrate multi-modal signals into LLMs, further bringing richer perception to entity agents and allowing embodied agents to perceive world-understanding tasks more delicately. However, existing works: 1) operate independently by agents, each containing multiple LLMs, from perception to action, resulting in gaps between complex tasks and execution; 2) train MLMs on static data, struggling with dynamics in open-ended scenarios; 3) input prior knowledge directly as prompts, suppressing application flexibility. We propose STEVE-2, a hierarchical knowledge distillation framework for open-ended embodied tasks, characterized by 1) a hierarchical system for multi-granular task division, 2) a mirrored distillation method for parallel simulation data, and 3) an extra expert model for bringing additional knowledge into parallel simulation. After distillation, embodied agents can complete complex, open-ended tasks without additional expert guidance, utilizing the performance and knowledge of a versatile MLM. Extensive evaluations on navigation and creation tasks highlight the superior performance of STEVE-2 in open-ended tasks, with $1.4 \times$ - $7.3 \times$ in performance.

Large language models (LLMs) have achieved impressive progress on several open-world tasks. Recently, using LLMs to build embodied agents has been a hotspot. In this paper, we propose STEVE, a comprehensive and visionary embodied agent in the Minecraft virtual environment. STEVE consists of three key components: vision perception, language instruction, and code action. Vision perception involves the interpretation of visual information in the environment, which is then integrated into the LLMs component with agent state and task instruction. Language instruction is responsible for iterative reasoning and decomposing complex tasks into manageable guidelines. Code action generates executable skill actions based on retrieval in skill database, enabling the agent to interact effectively within the Minecraft environment. We also collect STEVE-21K dataset, which includes 600$+$ vision-environment pairs, 20K knowledge question-answering pairs, and 200$+$ skill-code pairs. We conduct continuous block search, knowledge question and answering, and tech tree mastery to evaluate the performance. Extensive experiments show that STEVE achieves at most $1.5 \times$ faster unlocking key tech trees and $2.5 \times$ quicker in block search tasks compared to previous state-of-the-art methods.

Although 3D human pose estimation has gained impressive development in recent years, only a few works focus on infants, that have different bone lengths and also have limited data. Directly applying adult pose estimation models typically achieves low performance in the infant domain and suffers from out-of-distribution issues. Moreover, the limitation of infant pose data collection also heavily constrains the efficiency of learning-based models to lift 2D poses to 3D. To deal with the issues of small datasets, domain adaptation and data augmentation are commonly used techniques. Following this paradigm, we take advantage of an optimization-based method that utilizes generative priors to predict 3D infant keypoints from 2D keypoints without the need of large training data. We further apply a guided diffusion model to domain adapt 3D adult pose to infant pose to supplement small datasets. Besides, we also prove that our method, ZeDO-i, could attain efficient domain adaptation, even if only a small number of data is given. Quantitatively, we claim that our model attains state-of-the-art MPJPE performance of 43.6 mm on the SyRIP dataset and 21.2 mm on the MINI-RGBD dataset.

Learning-based methods have dominated the 3D human pose estimation (HPE) tasks with significantly better performance in most benchmarks than traditional optimization-based methods. Nonetheless, 3D HPE in the wild is still the biggest challenge for learning-based models, whether with 2D-3D lifting, image-to-3D, or diffusion-based methods, since the trained networks implicitly learn camera intrinsic parameters and domain-based 3D human pose distributions and estimate poses by statistical average. On the other hand, the optimization-based methods estimate results case-by-case, which can predict more diverse and sophisticated human poses in the wild. By combining the advantages of optimization-based and learning-based methods, we propose the \textbf{Ze}ro-shot \textbf{D}iffusion-based \textbf{O}ptimization (\textbf{ZeDO}) pipeline for 3D HPE to solve the problem of cross-domain and in-the-wild 3D HPE. Our multi-hypothesis \textit{\textbf{ZeDO}} achieves state-of-the-art (SOTA) performance on Human3.6M, with minMPJPE $51.4$mm, without training with any 2D-3D or image-3D pairs. Moreover, our single-hypothesis \textit{\textbf{ZeDO}} achieves SOTA performance on 3DPW dataset with PA-MPJPE $40.3$mm on cross-dataset evaluation, which even outperforms learning-based methods trained on 3DPW.