United States

Despite the advancements of Video Large Language Models (VideoLLMs) in various tasks, they struggle with fine-grained temporal understanding tasks, such as Dense Video Captioning (DVC). DVC is a complicated task of describing all events within a video while also temporally locating each event in a video, which integrates multiple fine-grained tasks, including video segmentation, video captioning, and temporal video grounding.
Previous VideoLLMs attempt to solve DVC in a single step, failing to utilize their reasoning capability.
Moreover, previous loss used for training VideoLLMs does not fully reflect evaluation metrics, therefore providing supervision not directly aligned to target tasks.
To address such a problem, we propose a novel framework named VidChain comprised of Chain-of-Tasks (CoTasks) and Metric-based Direct Preference Optimization (M-DPO).
CoTasks decompose a complex task into a sequence of sub-tasks, allowing VideoLLMs to leverage their reasoning capabilities more effectively.
M-DPO aligns a VideoLLM with evaluation metrics, providing fine-grained supervision to each task that is well-aligned with metrics.
Applied to two different VideoLLMs, VidChain consistently improves their fine-grained video understanding, thereby outperforming previous VideoLLMs on two different DVC benchmarks and also on the temporal video grounding task.

AAAI 2025

VidChain: Chain-of-Tasks with Metric-based Direct Preference Optimization for Dense Video Captioning

video understanding activity analysis

poster

We are pleased to announce the Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI-25), which will be held in Philadelphia, Pennsylvania at the Pennsylvania Convention Center from February 25 to March 4, 2025.

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-25 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.

### [Invited Speakers](https://aaai.org/conference/aaai/aaai-25/aaai-25-invited-speakers/)

Register [here](https://aaai.org/conference/aaai/aaai-25/registration/)

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-25 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.



Federated learning (FL) enables decentralized clients to collaboratively train a global model under the orchestration of a central server. However, the iterative exchange of model parameters between the server and clients imposes heavy communication burdens, risks potential privacy leakage, and even precludes collaboration among heterogeneous clients. Distillation-based FL tackles these challenges by exchanging low-dimensional model outputs rather than model parameters, yet it relies heavily on a task-relevant auxiliary dataset that is often not available in practice. Data-free FL attempts to overcome this limitation by training a server-side generator to directly synthesize task-specific data samples for knowledge transfer. However, the update rule of the generator requires clients to share on-device models for white-box access, which greatly compromises the advantages of distillation-based FL. This motivates us to explore a data-free and black-box FL framework via Zeroth-order Gradient Estimation (FedZGE), which estimates the gradients after flowing through on-device models in a black-box optimization manner to complete the training of the generator in terms of fidelity, transferability, diversity, and equilibrium, without involving any auxiliary data or sharing any model parameters, thus combining the advantages of both distillation-based FL and data-free FL. Experiments on large-scale image classification datasets and network architectures demonstrate the superiority of FedZGE in terms of data heterogeneity, model heterogeneity, communication efficiency, and privacy protection. Code is available at: https://anonymous.4open.science/r/FedZGE-B6A3.

Data-Free Black-Box Federated Learning via Zeroth-Order Gradient Estimation

Posters serve an essential function in marketing and advertising by improving visual communication and brand visibility, thus significantly contributing to industrial design. With the latest developments in controllable T2I diffusion models, research interest has surged in text rendering within synthesized images. Although text rendering accuracy has seen advancements, automatic poster generation remains a relatively untapped area. This paper presents an automatic poster generation framework featuring text rendering capabilities through the use of LLMs. Our framework employs a triple-cross attention mechanism based on alignment learning to achieve precise text placement within detailed contextual backgrounds. Moreover, it supports adjustable fonts, varying image resolutions, and poster rendering with textual prompts in both English and Chinese. Additionally, we present a comprehensive bilingual image-text dataset, GlyphDraw-3M, comprising 3 million image-text pairs, each with OCR annotations and resolutions exceeding 1024. Our method utilizes the SDXL architecture, and extensive experiments confirm its ability to generate posters with intricate and context-rich backgrounds.

GlyphDraw2: Automatic Generation of Complex Glyph Posters with Diffusion Models and Large Language Models

Large language models (LLMs) have significantly advanced the performance of various natural language processing tasks, including text-to-SQL. Current LLM-based text-to-SQL schemes mainly focus on improving the understanding of natural language questions (NLQs) or refining the quality of generated SQLs. While these strategies are effective, they often address specific, nuanced aspects. In contrast, humans approach text-to-SQL with a holistic view, applying transitional logical reasoning across multiple steps to arrive at the final answer. We believe LLMs can leverage human cognitive processes to achieve greater accuracy in text-to-SQL. In this paper, we present CogSQL, a framework featuring a suite of tailored models and strategies aimed at replicating human cognitive processes for enhanced LLM-based text-to-SQL. CogSQL consists of three key modules: (1) SQL preparation: we employ a coarse-to-fine schema linking and syntax keyword prediction, akin to how human recall and align key concepts for better understanding. (2) SQL generation: we introduce a concept-enhanced chain-of-thought prompting, enhancing NLQ interpretation and SQL composition of LLMs, similar to humans drafting SQL query. (3) SQL correction: we develop NLQ consistency and result consistency techniques to correct various errors, mirroring how humans evaluate and refine reasoning. We conduct extensive experiments using diverse benchmarks and LLMs. The results and analysis verify the effectiveness and generalizability of CogSQL.

CogSQL: A Cognitive Framework for Enhancing Large Language Models in Text-to-SQL Translation

Contrastive learning is a paradigm for learning representations from unlabelled data and several recent works have claimed that such models effectively learn spectral embeddings and show relations between (wide) contrastive models and kernel principal component analysis (PCA). However, it is not known if trained contrastive models indeed correspond to kernel methods or PCA. In this work, we analyze the training dynamics of two-layer contrastive models, with non-linear activation, and answer when these models are close to PCA or kernel methods. It is well known in the supervised setting that neural networks are equivalent to neural tangent kernel (NTK) machines, and that the NTK of infinitely wide networks remains constant during training. We provide the first constancy results of NTK for contrastive losses, and present a nuanced picture: NTK of wide networks remains almost constant for cosine similarity based contrastive losses, but not for losses based on dot product similarity. We further study the training dynamics of contrastive models with orthogonality constraints on output layer, which is implicitly assumed in works relating contrastive learning to spectral embedding. Our deviation bounds suggest that representations learned by contrastive models are close to the principal components of a certain matrix computed from random features.

When Can We Approximate Wide Contrastive Models with Neural Tangent Kernels and Principal Component Analysis?

Charts are widely used for data visualization across various fields, including education, research, and business. Chart Question Answering (CQA) is an emerging task focused on the automatic interpretation and reasoning of data presented in charts. However, chart images are inherently difficult to interpret, and chart-related questions often involve complex logical and numerical reasoning, which hinders the performance of existing models. This paper introduces VProChart, a novel framework designed to address these challenges in CQA by integrating a lightweight Visual Perception Alignment Agent (VPAgent) and a Programmatic Solution Reasoning approach. VPAgent aligns and models chart elements based on principles of human visual perception, enhancing the understanding of chart context. The Programmatic Solution Reasoning approach leverages large language models (LLMs) to transform natural language reasoning questions into structured solution programs, facilitating precise numerical and logical reasoning. Extensive experiments on benchmark datasets such as ChartQA and PlotQA demonstrate that VProChart significantly outperforms existing methods, highlighting its capability in understanding and reasoning with charts.

VProChart: Answering Chart Question Through Visual Perception Alignment Agent and Programmatic Solution Reasoning

Generative image watermarking inserts secret watermarks into generated images and plays an important role in tracing the usages of generative models. For watermarking of diffusion models, inversion-based framework emerges as an effective approach. 
Such framework employs a robust mechanism to embed the watermark into the starting latent before ``forward sampling'', thereby generating images with the implicit watermark. During watermark detection, inversion techniques are employed to reverse the process and obtain the watermarked latent, followed by further extraction. 
The robustness of this technique hinges primarily on the embedding mechanism and the inversion process. Previous methods predominantly focused on enhancing the robustness of the embedding mechanism but overlooked the reduction of the inversion errors. However, our results show that inversion error will significantly affect the overall robustness. Therefore, in this paper, we delve into the inversion error aspect and propose CoSDA, a \textbf{co}mpensation \textbf{s}ampling and \textbf{d}rift \textbf{a}lignment-based approach. The inversion error primarily accumulated during two stages: the internal error incurred by the algorithm, and the inevitable external noise. We observe that the main source of internal error comes from the mismatch in conditions (\textit{e.g.,} prompt, guidance scale) between forward and backward sampling processes. Therefore, we propose a compensation-based forward sampling, compensating for certain mismatch conditions and reducing the inversion error caused by the mismatch. Addressing external error caused by inevitable image distortions (\textit{e.g.,} JPEG compression), we introduce a drift-alignment approach, where a neural network is trained adversarially to restore the original watermarked latent from the distorted counterpart. Experimental results show that CoSDA effectively enhances watermark robustness while maintaining the visual quality of generated images.

CoSDA: Enhancing the Robustness of Inversion-based Generative Image Watermarking Framework

Concept Bottleneck Models (CBMs) offer inherent interpretability by initially translating images into human-comprehensible concepts, followed by a linear combination of these concepts for classification. However, the annotation of concepts for visual recognition tasks requires extensive expert knowledge and labor, constraining the broad adoption of CBMs. Recent approaches have leveraged the knowledge of large language models to construct concept bottlenecks, with multimodal models like CLIP subsequently mapping image features into the concept feature space for classification. Despite this, the concepts produced by language models can be verbose and may introduce non-visual attributes, which hurts accuracy and interpretability. In this study, we investigate to avoid these issues by constructing CBMs directly from multimodal models. To this end, we adopt common words as base concept vocabulary, and leverage auxiliary unlabeled images to construct a Vision-to-Concept (V2C) tokenizer that can explicitly quantize images into their most relevant visual concepts, thus creating a vision-oriented concept bottleneck tightly coupled with the multimodal model. This leads to our V2C-CBM that is training efficient and interpretable with high accuracy. Our V2C-CBM has matched or outperformed LLM-supervised CBMs on various visual classification benchmarks, validating the efficacy of our approach. The associated code will be released soon.

V2C-CBM: Building Concept Bottlenecks with Vision-to-Concept Tokenizer

Neural architecture search (NAS) enables finding the best-performing architecture from a search space automatically. Most NAS methods exploit an over-parameterized network (i.e., a supernet) containing all possible architectures (i.e., subnets) in the search space. However, the subnets that share the same set of parameters are likely to have different characteristics, interfering with each other during training. To address this, few-shot NAS methods have been proposed that divide the space into a few subspaces and employ a separate supernet for each subspace to limit the extent of weight sharing. They achieve state-of-the-art performance, but the computational cost increases accordingly. We introduce in this paper a novel few-shot NAS method that exploits the number of nonlinear functions to split the search space. To be specific, our method divides the space such that each subspace consists of subnets with the same number of nonlinear functions. Our splitting criterion is efficient, since it does not require comparing gradients of a supernet to split the space. In addition, we have found that dividing the space allows us to reduce the channel dimensions required for each supernet, which enables training multiple supernets in an efficient manner. We also introduce a supernet-balanced sampling (SBS) technique, sampling several subnets at each training step, to train different supernets evenly within a limited number of training steps. Extensive experiments on standard NAS benchmarks demonstrate the effectiveness of our approach. We will make our code and models publicly available online.

Efficient Few-Shot Neural Architecture Search by Counting the Number of Nonlinear Functions

Enabling Large Language Models (LLMs) to comprehend the 3D physical world remains a significant challenge. Due to the lack of large-scale 3D-text pair datasets, the success of LLMs has yet to be replicated in 3D understanding. In this paper, we rethink this issue and propose a new task: 3D Data-Efficient Point-Language Understanding. The goal is to enable LLMs to achieve robust 3D object understanding with minimal 3D point cloud and text data pairs. To address this task, we introduce GreenPLM, which leverages more text data to compensate for the lack of 3D data. First, inspired by using CLIP to align images and text, we utilize a pre-trained point cloud-text encoder to map the 3D point cloud space to the text space. This mapping leaves us to seamlessly connect the text space with LLMs. Once the point-text-LLM connection is established, we further enhance text-LLM alignment by expanding the intermediate text space, thereby reducing the reliance on 3D point cloud data. Specifically, we generate 6M free-text descriptions of 3D objects, and design a three-stage training strategy to help LLMs better explore the intrinsic connections between different modalities. To achieve efficient modality alignment, we design a zero-parameter cross-attention module for token pooling. Extensive experimental results show that GreenPLM requires only 12\% of the 3D training data used by existing state-of-the-art models to achieve superior 3D understanding. Remarkably, GreenPLM also achieves competitive performance using text-only data. We will release the code and weights after review.

More Text,  Less Point: Towards 3D Data-Efficient Point-Language Understanding

Unsupervised point cloud shape correspondence aims to establish dense correspondences between source and target point clouds. Existing methods universally follow a one-step paradigm where point-by-point similarities are utilized to directly obtain shape correspondence. The one-step paradigm shows satisfactory performance for shapes with minor motions, but it often fails in large-scale motions of humans and animals. To address this challenge, we propose a conditional Diffusion model with reliable pseudo-label guidance for unsupervised point cloud shape Correspondence (DiffCorr), including a transformer-based conditional diffusion model and a reliable pseudo-label generator. The proposed DiffCorr enjoys several merits. Firstly, the transformer-based conditional diffusion model implements a coarse-to-fine optimization for coarse correspondences conditioned on the initial correspondences and the local structures. Secondly, we design a reliable pseudo-label generator to provide high-quality pseudo-labels for training the diffusion model. Extensive experiments on four human and animal datasets demonstrate that DiffCorr surpasses state-of-the-art unsupervised point cloud shape correspondence methods and exhibits favorable generalization capabilities.

VidChain: Chain-of-Tasks with Metric-based Direct Preference Optimization for Dense Video Captioning

Downloads

Next from AAAI 2025

Data-Free Black-Box Federated Learning via Zeroth-Order Gradient Estimation

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES