LeMo: Enabling LEss Token Involvement for MOre Context Fine-tuning

The escalating demand for long-context applications has intensified the necessity of extending the LLM context windows. Despite recent fine-tuning approaches successfully expanding context lengths, their high memory footprints, especially for activations, present a critical practical limitation. Current parameter-efficient fine-tuning methods prioritize reducing parameter update overhead over addressing activation memory constraints. Similarly, existing sparsity mechanisms improve computational efficiency but overlook activation memory optimization due to the phenomenon of Shadowy Activation. In this paper, we propose LeMo, the first LLM fine-tuning system that explores and exploits a new token-level sparsity mechanism inherent in long-context scenarios, termed Contextual Token Sparsity. LeMo minimizes redundant token involvement by assessing the informativeness of token embeddings while preserving model accuracy. Specifically, LeMo introduces three key techniques: (1) Token Elimination, dynamically identifying and excluding redundant tokens across varying inputs and layers. (2) Pattern Prediction, utilizing well-trained predictors to approximate token sparsity patterns with minimal overhead. (3) Kernel Optimization, employing permutation-free and segment-based strategies to boost system performance. We implement LeMo as an end-to-end fine-tuning system compatible with various LLM architectures and other optimization techniques. Comprehensive evaluations demonstrate that LeMo reduces memory consumption by up to 1.93x and achieves up to 1.36x speedups, outperforming state-of-the-art fine-tuning systems.

Enhancing Low-Cost Video Editing with Lightweight Adaptors and Temporal-Aware Inversion

Recent advancements in text-to-image (T2I) generation using diffusion models have enabled cost-effective video-editing applications by leveraging pre-trained models, eliminating the need for resource-intensive training. However, the frame-independence of T2I generation often results in poor temporal consistency. Existing methods address this issue through temporal layer fine-tuning or inference-based temporal propagation, but these approaches suffer from high training costs or limited temporal coherence. To address these challenges, we propose a General and Efficient Adapter (GE-Adapter) that integrates temporal-spatial and semantic consistency with Baliteral DDIM inversion. This framework introduces three key components: (1) Frame-based Temporal Consistency Blocks (FTC Blocks) to capture frame-specific features and enforce smooth inter-frame transitions via temporally-aware loss functions; (2) Channel-dependent Spatial Consistency Blocks (SCD Blocks) employing bilateral filters to enhance spatial coherence by reducing noise and artifacts; and (3) Token-based Semantic Consistency Module (TSC Module) to maintain semantic alignment using shared prompt tokens and frame-specific tokens. Our method significantly improves perceptual quality, text-image alignment, and temporal coherence, as demonstrated on the MSR-VTT dataset. Additionally, it achieves enhanced fidelity and frame-to-frame coherence, offering a practical solution for T2V editing.

Scale-wise Bidirectional Alignment Network for Referring Remote Sensing Image Segmentation

The goal of referring remote sensing image segmentation (RRSIS) is to extract specific pixel-level regions within an aerial image via a natural language expression. Recent advancements, particularly Transformer-based fusion designs, have demonstrated remarkable progress in this domain. However, existing methods primarily focus on refining visual features using language-aware guidance during the cross-modal fusion stage, neglecting the complementary vision-to-language flow. This limitation often leads to irrelevant or suboptimal representations. In addition, the diverse spatial scales of ground objects in aerial images pose significant challenges to the visual perception capabilities of existing models when conditioned on textual inputs. In this paper, we propose an innovative framework called Scale-wise Bidirectional Alignment Network (SBANet) to address these challenges for RRSIS. Specifically, we design a Bidirectional Alignment Module (BAM) with learnable query tokens to selectively and effectively represent visual and linguistic features, emphasizing regions associated with key tokens. BAM is further enhanced with a dynamic feature selection block, designed to provide both macro- and micro-level visual features, preserving global context and local details to facilitate more effective cross-modal interaction. Furthermore, SBANet incorporates a text-conditioned channel and spatial aggregator to bridge the gap between the encoder and decoder, enhancing cross-scale information exchange in complex aerial scenarios. Extensive experiments demonstrate that our proposed method achieves superior performance in comparison to previous state-of-the-art methods on the RRSIS-D and RefSegRS datasets, both quantitatively and qualitatively. The code will be released after publication.

Temporal-Frequency State Space Duality: An Efficient Paradigm for Speech Emotion Recognition

Speech Emotion Recognition (SER) plays a critical role in enhancing user experience within human-computer interaction. However, existing methods are overwhelmed by temporal domain analysis, overlooking the valuable envelope structures of the frequency domain that are equally important for robust emotion recognition. To overcome this limitation, we propose TF-Mamba, a novel multi-domain framework that captures emotional expressions in both temporal and frequency dimensions.Concretely, we propose a temporal-frequency mamba block to extract temporal- and frequency-aware emotional features, achieving an optimal balance between computational efficiency and model expressiveness. Besides, we design a Complex Metric-Distance Triplet (CMDT) loss to enable the model to capture representative emotional clues for SER. Extensive experiments on the IEMOCAP and MELD datasets show that TF-Mamba surpasses existing methods in terms of model size and latency, providing a more practical solution for future SER applications.

Prototypical Calibrating Ambiguous Samples for Micro-Action Recognition

Micro-Action Recognition (MAR) has gained increasing attention due to its crucial role as a form of non-verbal communication in social interactions, with promising potential for applications in human communication and emotion analysis. However, current approaches often overlook the inherent ambiguity in micro-actions, which arises from the wide category range and subtle visual differences between categories. This oversight hampers the accuracy of micro-action recognition. In this paper, we propose a novel Prototypical Calibrating Ambiguous Network (\textbf{PCAN}) to unleash and mitigate the ambiguity of MAR. \textbf{Firstly}, we employ a hierarchical action-tree to identify the ambiguous sample, categorizing them into distinct sets of ambiguous samples of false negatives and false positives, considering both body- and action-level categories. \textbf{Secondly}, we implement an ambiguous contrastive refinement module to calibrate these ambiguous samples by regulating the distance between ambiguous samples and their corresponding prototypes. This calibration process aims to pull false negative ($\mathbb{FN}$) samples closer to their respective prototypes and push false positive ($\mathbb{FP}$) samples apart from their affiliated prototypes. In addition, we propose a new prototypical diversity amplification loss to strengthen the model's capacity by amplifying the differences between different prototypes. \textbf{Finally}, we propose a prototype-guided rectification to rectify prediction by incorporating the representability of prototypes. Extensive experiments conducted on the benchmark dataset demonstrate the superior performance of our method compared to existing approaches. The code is available at https://github.com/kunli-cs/PCAN.

Patch-level Sounding Object Tracking for Audio-Visual Question Answering

Answering questions related to audio-visual scenes, i.e., the AVQA task, is becoming increasingly popular. A critical challenge is accurately identifying and tracking sounding objects related to the question along the timeline. In this paper, we present a new Patch-level Sounding Object Tracking (PSOT) method. It begins with a Motion-driven Key Patch Tracking (M-KPT) module, which relies on visual motion information to identify salient visual patches with significant movements that are more likely to relate to sounding objects and questions. We measure the patch-wise motion intensity map between neighboring video frames and utilize it to construct and guide a motion-driven graph network. Meanwhile, we design a Sound-driven KPT (S-KPT) module to explicitly track sounding patches. This module also involves a graph network, with the adjacency matrix regularized by the audio-visual correspondence map. The M-KPT and S-KPT modules are performed in parallel for each temporal segment, allowing balanced tracking of salient and sounding objects. Based on the tracked patches, we further propose a Question-driven KPT (Q-KPT) module to retain patches highly relevant to the question, ensuring the model focuses on the most informative clues. The audio-visual-question features are updated during the processing of these modules, which are then aggregated for final answer prediction. Extensive experiments on standard datasets demonstrate the effectiveness of our method, achieving competitive performance even compared to recent large-scale pretraining-based approaches.

GPTDrawer: Enhancing Visual Synthesis through ChatGPT

In the burgeoning field of AI-driven image generation, the quest for precision and relevance in response to textual prompts remains paramount. This paper introduces GPTDrawer, an innovative pipeline that leverages the generative prowess of GPT-based models to enhance the visual synthesis process. Our methodology employs a novel algorithm that iteratively refines input prompts using keyword extraction, semantic analysis, and image-text congruence evaluation. By integrating ChatGPT for natural language processing and Stable Diffusion for image generation, GPTDrawer produces a batch of images that undergo successive refinement cycles, guided by cosine similarity metrics until a threshold of semantic alignment is attained. The results demonstrate a marked improvement in the fidelity of images generated in accordance with user-defined prompts, showcasing the system's ability to interpret and visualize complex semantic constructs. The implications of this work extend to various applications, from creative arts to design automation, setting a new benchmark for AI-assisted creative processes.

Repetitive Action Counting with Hybrid Temporal Relation Modeling

Repetitive Action Counting (RAC) aims to count the number of repetitive actions occurring in videos. In the real world, repetitive actions have great diversity and bring numerous challenges (e.g., viewpoint changes, non-uniform periods, and action interruptions). Existing methods based on the temporal self-similarity matrix (TSSM) for RAC are trapped in the bottleneck of insufficient capturing action periods when applied to complicated daily videos. To tackle this issue, we propose a novel method named Hybrid Temporal Relation Modeling Network (HTRM-Net) to build diverse TSSM for RAC. The HTRM-Net mainly consists of three key components: bi-modal temporal self-similarity matrix modeling, random matrix dropping, and local temporal context modeling. Specifically, we construct temporal self-similarity matrices by bi-modal (self-attention and dual-softmax) operations, yielding diverse matrix representations from the combination of row-wise and column-wise correlations. To further enhance matrix representations, we propose incorporating a random matrix dropping module to guide channel-wise learning of the matrix explicitly. After that, we inject the local temporal context of video frames and the learned matrix into temporal correlation modeling, which can make the model robust enough to cope with error-prone situations, such as action interruption. Finally, a multi-scale matrix fusion module is designed to aggregate temporal correlations adaptively in multi-scale matrices. Extensive experiments across intra- and cross-datasets demonstrate that the proposed method not only outperforms current state-of-the-art methods but also exhibits robust capabilities in accurately counting repetitive actions in unseen action categories. Notably, our method surpasses the classical TransRAC method by 20.04\% in MAE and 22.76\% in OBO.

FOF-X: Towards Real-time Detailed Human Reconstruction from a Single Image

We introduce FOF-X for real-time reconstruction of detailed human geometry from a single image. Balancing real-time speed against high-quality results is a persistent challenge, mainly due to the high computational demands of existing 3D representations. To address this, we propose Fourier Occupancy Field (FOF), an efficient 3D representation by learning the Fourier series. The core of FOF is to factorize a 3D occupancy field into a 2D vector field, retaining topology and spatial relationships within the 3D domain while facilitating compatibility with 2D convolutional neural networks. Such a representation bridges the gap between 3D and 2D domains, enabling the integration of human parametric models as priors and enhancing the reconstruction robustness. Based on FOF, we design a new reconstruction framework, FOF-X, to avoid the performance degradation caused by texture and lighting. This enables our real-time reconstruction system to better handle the domain gap between training images and real images. Additionally, in FOF-X, we enhance the inter-conversion algorithms between FOF and mesh representations with a Laplacian constraint and an automaton-based discontinuity matcher, improving both quality and robustness. We validate the strengths of our approach on different datasets and real-captured data, where FOF-X achieves new state-of-the-art results. The code will be released for research purposes.

Fine-Grained Behavior Simulation with Role-Playing Large Language Model on Social Media

Large language models (LLMs) have demonstrated impressive capabilities in role-playing tasks. However, there is limited research on whether LLMs can accurately simulate user behavior in real-world scenarios, such as social media. This requires models to effectively analyze a user's history and simulate their role. In this paper, we introduce \textbf{FineRob}, a novel fine-grained behavior simulation dataset. We collect the complete behavioral history of 1,866 distinct users across three social media platforms. Each behavior is decomposed into three fine-grained elements: object, type, and content, resulting in 78.6k QA records. Based on FineRob, we identify two dominant reasoning patterns in LLMs' behavior simulation processes and propose the \textbf{OM-CoT} fine-tuning method to enhance the capability. Through comprehensive experiments, we conduct an in-depth analysis of key factors of behavior simulation and also demonstrate the effectiveness of OM-CoT approach\footnote{Code and dataset are available at \url{https://github.com/linkseed18612254945/FineRob}}