arXiv subjects

Danfeng Yan

Publications and source records attributed to Danfeng Yan.

8 recordsLinked to original sources

arXiv2026

FADE: From Passive Verification to Active Discovery in Counterfactual Video Understanding

Counterfactual video understanding evaluates whether models grasp physical and commonsense regularities. However, existing multiple-choice question (MCQ) benchmarks inadvertently leak target events through their questions and candidate options. This reduces the core challenge from active discovery to text-guided verification. In this paper, we present FADE, an effective training framework for counterfactual discovery and explanation. Our method is built on an evidence-first, two-stage training paradigm. First, evidence-internalized supervised fine-tuning grounds the model's predictions in decisive visual anomalies. Second, we apply a fading-anchor reinforcement learning strategy that progressively removes textual guidance, compelling the model to independently discover and explain evidence. To rigorously evaluate this capability, we also introduce an effective pipeline that converts existing MCQ datasets into aligned MCQ, open-ended question answering (OQA), and captioning tasks without requiring additional data curation. Our simple approach yields strong results. Using Qwen3-VL-8B as the baseline, FADE achieves state-of-the-art strict paired scores across all three tasks on DualityVidQA-test and IPV-Bench, outperforming GPT-5.6. In specific, when transitioning from constrained MCQs to unconstrained OQA and captioning, our model demonstrates remarkable robustness. Its performance retention is 90.4% and 67.4% on DualityVidQA-test-substantially higher than the 48.1% and 30.7% retained by GPT-5.6. We hope this simple framework can serve as a solid baseline for future research in unconstrained counterfactual video understanding.

cs.CV↗

arXiv2025

VideoPerceiver: Enhancing Fine-Grained Temporal Perception in Video Multimodal Large Language Models

We propose VideoPerceiver, a novel video multimodal large language model (VMLLM) that enhances fine-grained perception in video understanding, addressing VMLLMs' limited ability to reason about brief actions in short clips or rare transient events in long videos. VideoPerceiver adopts a two-stage training framework. During supervised fine-tuning (SFT), we construct "key-information-missing" videos by extracting event-action keywords from captions, identifying corresponding key frames, and replacing them with adjacent frames. We jointly encode original and modified video tokens with text tokens, aligning intermediate visual representations with keywords via an auxiliary contrastive loss to enhance sensitivity to fine-grained motion cues. In reinforcement learning (RL), both video variants are fed into the model to generate descriptions, and a novel relative reward ensures responses from complete videos outperform those from degraded inputs, explicitly training the model to recover temporally precise action details. We also curate a dataset of 80,000 videos with fine-grained actions and transient events. Experiments show VideoPerceiver substantially outperforms state-of-the-art VMLLMs on fine-grained action understanding and rare event captioning benchmarks, while maintaining strong performance on standard tasks. By prioritizing task-relevant visual features, our work redefines video-language model training for fine-grained perception.

cs.CV↗

arXiv2025

SitLLM: Large Language Models for Sitting Posture Health Understanding via Pressure Sensor Data

Poor sitting posture is a critical yet often overlooked factor contributing to long-term musculoskeletal disorders and physiological dysfunctions. Existing sitting posture monitoring systems, although leveraging visual, IMU, or pressure-based modalities, often suffer from coarse-grained recognition and lack the semantic expressiveness necessary for personalized feedback. In this paper, we propose \textbf{SitLLM}, a lightweight multimodal framework that integrates flexible pressure sensing with large language models (LLMs) to enable fine-grained posture understanding and personalized health-oriented response generation. SitLLM comprises three key components: (1) a \textit{Gaussian-Robust Sensor Embedding Module} that partitions pressure maps into spatial patches and injects local noise perturbations for robust feature extraction; (2) a \textit{Prompt-Driven Cross-Modal Alignment Module} that reprograms sensor embeddings into the LLM's semantic space via multi-head cross-attention using the pre-trained vocabulary embeddings; and (3) a \textit{Multi-Context Prompt Module} that fuses feature-level, structure-level, statistical-level, and semantic-level contextual information to guide instruction comprehension.

cs.CL↗

arXiv2025

FaVChat: Hierarchical Prompt-Query Guided Facial Video Understanding with Data-Efficient GRPO

Existing video large language models (VLLMs) primarily leverage prompt agnostic visual encoders, which extract untargeted facial representations without awareness of the queried information, leading to the loss of task critical cues. To address this challenge, we propose FaVChat, the first VLLM designed for reasoning over subtle visual and dynamic facial cues. FaVChat introduces a hierarchical, prompt guided visual feature extraction framework that emphasizes question relevant information at three complementary levels. These multi level features are dynamically fused and injected into the LLM, enabling more accurate facial details reasoning To further improve learning efficiency under data scarcity, we propose Data Efficient GRPO, a reinforcement learning strategy that iteratively identifies high utility samples and maximizes the contribution of each instance via per instance utility estimation, substantially enhancing performance gains under limited supervision. We construct a large scale benchmark dataset FaVChat 170K, comprising approximately 60K high quality facial videos and 170K question answer pairs focusing on fine grained facial details. Extensive experiments, including zero shot evaluations on four facial understanding tasks, demonstrate that FaVChat consistently outperforms existing VLLMs.

cs.CV↗

arXiv2025

Knowledge Distillation for Image Restoration : Simultaneous Learning from Degraded and Clean Images

Model compression through knowledge distillation has seen extensive application in classification and segmentation tasks. However, its potential in image-to-image translation, particularly in image restoration, remains underexplored. To address this gap, we propose a Simultaneous Learning Knowledge Distillation (SLKD) framework tailored for model compression in image restoration tasks. SLKD employs a dual-teacher, single-student architecture with two distinct learning strategies: Degradation Removal Learning (DRL) and Image Reconstruction Learning (IRL), simultaneously. In DRL, the student encoder learns from Teacher A to focus on removing degradation factors, guided by a novel BRISQUE extractor. In IRL, the student decoder learns from Teacher B to reconstruct clean images, with the assistance of a proposed PIQE extractor. These strategies enable the student to learn from degraded and clean images simultaneously, ensuring high-quality compression of image restoration models. Experimental results across five datasets and three tasks demonstrate that SLKD achieves substantial reductions in FLOPs and parameters, exceeding 80\%, while maintaining strong image restoration performance.

cs.CV↗

arXiv2025

Soft Knowledge Distillation with Multi-Dimensional Cross-Net Attention for Image Restoration Models Compression

Transformer-based encoder-decoder models have achieved remarkable success in image-to-image transfer tasks, particularly in image restoration. However, their high computational complexity-manifested in elevated FLOPs and parameter counts-limits their application in real-world scenarios. Existing knowledge distillation methods in image restoration typically employ lightweight student models that directly mimic the intermediate features and reconstruction results of the teacher, overlooking the implicit attention relationships between them. To address this, we propose a Soft Knowledge Distillation (SKD) strategy that incorporates a Multi-dimensional Cross-net Attention (MCA) mechanism for compressing image restoration models. This mechanism facilitates interaction between the student and teacher across both channel and spatial dimensions, enabling the student to implicitly learn the attention matrices. Additionally, we employ a Gaussian kernel function to measure the distance between student and teacher features in kernel space, ensuring stable and efficient feature learning. To further enhance the quality of reconstructed images, we replace the commonly used L1 or KL divergence loss with a contrastive learning loss at the image level. Experiments on three tasks-image deraining, deblurring, and denoising-demonstrate that our SKD strategy significantly reduces computational complexity while maintaining strong image restoration capabilities.

cs.CV↗

arXiv2024

reCSE: Portable Reshaping Features for Sentence Embedding in Self-supervised Contrastive Learning

We propose reCSE, a self supervised contrastive learning sentence representation framework based on feature reshaping. This framework is different from the current advanced models that use discrete data augmentation methods, but instead reshapes the input features of the original sentence, aggregates the global information of each token in the sentence, and alleviates the common problems of representation polarity and GPU memory consumption linear increase in current advanced models. In addition, our reCSE has achieved competitive performance in semantic similarity tasks. And the experiment proves that our proposed feature reshaping method has strong universality, which can be transplanted to other self supervised contrastive learning frameworks and enhance their representation ability, even achieving state-of-the-art performance. Our code is available at https://github.com/heavenhellchen/reCSE.

cs.CL↗

arXiv2023

Uni-Removal: A Semi-Supervised Framework for Simultaneously Addressing Multiple Degradations in Real-World Images

Removing multiple degradations, such as haze, rain, and blur, from real-world images poses a challenging and illposed problem. Recently, unified models that can handle different degradations have been proposed and yield promising results. However, these approaches focus on synthetic images and experience a significant performance drop when applied to realworld images. In this paper, we introduce Uni-Removal, a twostage semi-supervised framework for addressing the removal of multiple degradations in real-world images using a unified model and parameters. In the knowledge transfer stage, Uni-Removal leverages a supervised multi-teacher and student architecture in the knowledge transfer stage to facilitate learning from pretrained teacher networks specialized in different degradation types. A multi-grained contrastive loss is introduced to enhance learning from feature and image spaces. In the domain adaptation stage, unsupervised fine-tuning is performed by incorporating an adversarial discriminator on real-world images. The integration of an extended multi-grained contrastive loss and generative adversarial loss enables the adaptation of the student network from synthetic to real-world domains. Extensive experiments on real-world degraded datasets demonstrate the effectiveness of our proposed method. We compare our Uni-Removal framework with state-of-the-art supervised and unsupervised methods, showcasing its promising results in real-world image dehazing, deraining, and deblurring simultaneously.

cs.CV↗