Captions Speak Louder than Images (CASLIE): Generalizing Foundation Models for E-commerce from High-quality Multimodal Instruction Data

Leveraging multimodal data to drive breakthroughs in e-commerce applications through Multimodal Foundation Models (MFMs) is gaining increasing attention from the research community. However, there are significant challenges that hinder the optimal use of multimodal e-commerce data by foundation models: (1) the scarcity of large-scale, high-quality multimodal benchmark datasets; and (2) the lack of effective multimodal information integration methods. To address these challenges, in this paper, we introduce MMECInstruct, the first-ever, large-scale, and high-quality multimodal instruction dataset for e-commerce. We also develop CASLIE, a simple, lightweight, yet effective framework for integrating multimodal information for e-commerce. Leveraging MMECInstruct, we fine-tune a series of e-commerce MFMs within CASLIE, denoted as CASLIE models. Our comprehensive evaluation demonstrates that CASLIE models substantially outperform 5 categories of advanced baseline models in the in-domain evaluation. Moreover, CASLIE models show strong generalizability to out-of-domain settings. MMECInstruct and CASLIE models are publicly accessible through https://ninglab.github.io/CASLIE/.

SAPIENT: Mastering Multi-turn Conversational Recommendation with Strategic Planning and Monte Carlo Tree Search

Conversational Recommender Systems (CRS) proactively engage users in interactive dialogues to elicit user preferences and provide personalized recommendations. Existing methods train Reinforcement Learning (RL)-based agent with greedy action selection or sampling strategy, and may suffer from suboptimal conversational planning. To address this, we present a novel Monte Carlo Tree Search (MCTS)-based CRS framework SAPIENT. SAPIENT consists of a conversational agent (S-agent) and a conversational planner (S-planner). S-planner builds a conversational search tree with MCTS based on the initial actions proposed by S-agent to find conversation plans. The best conversation plans from S-planner are used to guide the training of S-agent, creating a self-training loop where S-agent can iteratively improve its capability for conversational planning. Furthermore, we propose an efficient variant SAPIENT-e for trade-off between training efficiency and performance. Extensive experiments on four benchmark datasets validate the effectiveness of our approach, showing that SAPIENT outperforms the state-of-the-art baselines.

Dark Miner: Defend against unsafe generation for text-to-image diffusion models

Text-to-image diffusion models have been demonstrated with unsafe generation due to unfiltered large-scale training data, such as violent, sexual, and shocking images, necessitating the erasure of unsafe concepts. Most existing methods focus on modifying the generation probabilities conditioned on the texts containing unsafe descriptions. However, they fail to guarantee safe generation for unseen texts in the training phase, especially for the prompts from adversarial attacks. In this paper, we re-analyze the erasure task and point out that existing methods cannot guarantee the minimization of the total probabilities of unsafe generation. To tackle this problem, we propose Dark Miner. It entails a recurring three-stage process that comprises mining, verifying, and circumventing. It greedily mines embeddings with maximum generation probabilities of unsafe concepts and reduces unsafe generation more effectively. In the experiments, we evaluate its performance on two inappropriate concepts, two objects, and two styles. Compared with 6 previous state-of-the-art methods, our method achieves better erasure and defense results in most cases, especially under 4 state-of-the-art attacks, while preserving the model's native generation capability. Our code will be available on GitHub.

FSL-LVLM: Friction-Aware Safety Locomotion using Large Vision Language Model in Wheeled Robots

Wheeled-legged robots offer significant mobility and versatility but face substantial challenges when operating on slippery terrains. Traditional model-based controllers for these robots assume no slipping. While reinforcement learning (RL) helps quadruped robots adapt to different surfaces, recovering from slips remains challenging, especially for systems with few contact points. Estimating the ground friction coefficient is another open challenge. In this paper, we propose a novel friction-aware safety locomotion framework that integrates Large Vision Language Models (LVLMs) with a RL policy. Our approach explicitly incorporates the estimated friction coefficient into the RL policy, enabling the robot to adapt its behavior in advance based on the surface type before reaching it. We introduce a Friction-From-Vision (FFV) module, which leverages LVLMs to estimate ground friction coefficients, eliminating the need for large datasets and extensive training. The framework was validated on a customized wheeled inverted pendulum, and experimental results demonstrate that our framework increases the success rate in completing driving tasks by adjusting speed according to terrain type, while achieving better tracking performance compared to baseline methods. Our framework can be simply integrated with any other RL policies.

Generating Event-oriented Attribution for Movies via Two-Stage Prefix-Enhanced Multimodal LLM

The prosperity of social media platforms has raised the urgent demand for semantic-rich services, e.g., event and storyline attribution. However, most existing research focuses on clip-level event understanding, primarily through basic captioning tasks, without analyzing the causes of events across an entire movie. This is a significant challenge, as even advanced multimodal large language models (MLLMs) struggle with extensive multimodal information due to limited context length. To address this issue, we propose a Two-Stage Prefix-Enhanced MLLM (TSPE) approach for event attribution, i.e., connecting associated events with their causal semantics, in movie videos. In the local stage, we introduce an interaction-aware prefix that guides the model to focus on the relevant multimodal information within a single clip, briefly summarizing the single event. Correspondingly, in the global stage, we strengthen the connections between associated events using an inferential knowledge graph, and design an event-aware prefix that directs the model to focus on associated events rather than all preceding clips, resulting in accurate event attribution. Comprehensive evaluations of two real-world datasets demonstrate that our framework outperforms state-of-the-art methods.

VI3DRM:Towards meticulous 3D Reconstruction from Sparse Views via Photo-Realistic Novel View Synthesis

Recently, methods like Zero-1-2-3 have focused on single-view based 3D reconstruction and have achieved remarkable success. However, their predictions for unseen areas heavily rely on the inductive bias of large-scale pretrained diffusion models. Although subsequent work, such as DreamComposer, attempts to make predictions more controllable by incorporating additional views, the results remain unrealistic due to feature entanglement in the vanilla latent space, including factors such as lighting, material, and structure. To address these issues, we introduce the Visual Isotropy 3D Reconstruction Model (VI3DRM), a diffusion-based sparse views 3D reconstruction model that operates within an ID consistent and perspective-disentangled 3D latent space. By facilitating the disentanglement of semantic information, color, material properties and lighting, VI3DRM is capable of generating highly realistic images that are indistinguishable from real photographs. By leveraging both real and synthesized images, our approach enables the accurate construction of pointmaps, ultimately producing finely textured meshes or point clouds. On the NVS task, tested on the GSO dataset, VI3DRM significantly outperforms state-of-the-art method DreamComposer, achieving a PSNR of 38.61, an SSIM of 0.929, and an LPIPS of 0.027. Code will be made available upon publication.

CLIP-AGIQA: Boosting the Performance of AI-Generated Image Quality Assessment with CLIP

With the rapid development of generative technologies, AI-Generated Images (AIGIs) have been widely applied in various aspects of daily life. However, due to the immaturity of the technology, the quality of the generated images varies, so it is important to develop quality assessment techniques for the generated images. Although some models have been proposed to assess the quality of generated images, they are inadequate when faced with the ever-increasing and diverse categories of generated images. Consequently, the development of more advanced and effective models for evaluating the quality of generated images is urgently needed. Recent research has explored the significant potential of the visual language model CLIP in image quality assessment, finding that it performs well in evaluating the quality of natural images. However, its application to generated images has not been thoroughly investigated. In this paper, we build on this idea and further explore the potential of CLIP in evaluating the quality of generated images. We design CLIP-AGIQA, a CLIP-based regression model for quality assessment of generated images, leveraging rich visual and textual knowledge encapsulated in CLIP. Particularly, we implement multi-category learnable prompts to fully utilize the textual knowledge in CLIP for quality assessment. Extensive experiments on several generated image quality assessment benchmarks, including AGIQA-3K and AIGCIQA2023, demonstrate that CLIP-AGIQA outperforms existing IQA models, achieving excellent results in evaluating the quality of generated images.

S^3D-NeRF: Single-Shot Speech-Driven Neural Radiance Field for High Fidelity Talking Head Synthesis

Talking head synthesis is a practical technique with wide applications. Current Neural Radiance Field (NeRF) based approaches have shown their superiority on driving one-shot talking heads with videos or signals regressed from audio. However, most of them failed to take the audio as driven information directly, unable to enjoy the flexibility and availability of speech. Since mapping audio signals to face deformation is non-trivial, we design a Single-Shot Speech-Driven Neural Radiance Field (S^3D-NeRF) method in this paper to tackle the following three difficulties: learning a representative appearance feature for each identity, modeling motion of different face regions with audio, and keeping the temporal consistency of the lip area. To this end, we introduce a Hierarchical Facial Appearance Encoder to learn multi-scale representations for catching the appearance of different speakers, and elaborate a Cross-modal Facial Deformation Field to perform speech animation according to the relationship between the audio signal and different face regions. Moreover, to enhance the temporal consistency of the important lip area, we introduce a lip-sync discriminator to penalize the out-of-sync audio-visual sequences. Extensive experiments have shown that our S^3D-NeRF surpasses previous arts on both video fidelity and audio-lip synchronization.

SWaT: Statistical Modeling of Video Watch Time through User Behavior Analysis

The significance of estimating video watch time has been highlighted by the rising importance of (short) video recommendation, which has become a core product of mainstream social media platforms. Modeling video watch time, however, has been challenged by the complexity of user-video interaction, such as different user behavior modes in watching the recommended videos and varying watching probabilities over the video horizon. Despite the importance and challenges, existing literature on modeling video watch time mostly focuses on relatively black-box mechanical enhancement of the classical regression/classification losses, without factoring in user behavior in a principled manner. In this paper, we for the first time take on a user-centric perspective to model video watch time, from which we propose a white-box statistical framework that directly translates various user behavior assumptions in watching (short) videos into statistical watch time models. These behavior assumptions are portrayed by our domain knowledge on users' behavior modes in video watching. We further employ bucketization to cope with user's non-stationary watching probability over the video horizon, which additionally helps to respect the constraint of video length and facilitate the practical compatibility between the continuous regression event of watch time and other binary classification events. We test our models extensively on two public datasets, a large-scale offline industrial dataset, and an online A/B test on a short video platform with hundreds of millions of daily-active users. On all experiments, our models perform competitively against strong relevant baselines, demonstrating the efficacy of our user-centric perspective and proposed framework.

TextIM: Part-aware Interactive Motion Synthesis from Text

In this work, we propose TextIM, a novel framework for synthesizing TEXT-driven human Interactive Motions, with a focus on the precise alignment of part-level semantics. Existing methods often overlook the critical roles of interactive body parts and fail to adequately capture and align part-level semantics, resulting in inaccuracies and even erroneous movement outcomes. To address these issues, TextIM utilizes a decoupled conditional diffusion framework to enhance the detailed alignment between interactive movements and corresponding semantic intents from textual descriptions. Our approach leverages large language models, functioning as a human brain, to identify interacting human body parts and to comprehend interaction semantics to generate complicated and subtle interactive motion. Guided by the refined movements of the interacting parts, TextIM further extends these movements into a coherent whole-body motion. We design a spatial coherence module to complement the entire body movements while maintaining consistency and harmony across body parts using a part graph convolutional network. For training and evaluation, we carefully selected and re-labeled interactive motions from HUMANML3D to develop a specialized dataset. Experimental results demonstrate that TextIM produces semantically accurate human interactive motions, significantly enhancing the realism and applicability of synthesized interactive motions in diverse scenarios, even including interactions with deformable and dynamically changing objects.