With the explosion of human-machine interaction, emotion recognition has reignited attention. Previous works focus on improving visual feature fusion and reasoning from multiple image levels. Although it is non-trivial to deduce a person’s emotion by integrating multi-level feature (head, body and context), the emotion recognition results of each level is usually different from one another, which creates inconsistency in the prevailing feature alignment method and decrease recognition performance. In this work, we propose a multi-level image feature refinement method for emotion recognition (CocoER) to mitigate the impact caused by conflicting results from multi-level recognition. First, we leverage cross-level attention to improve visual feature consistency between hierarchically cropped head, body and context windows. Then, vocabulary informed alignment is incorporated into the recognition framework to produce pseudo label and guide hierarchical visual feature refinement. To effectively fuse multi-level feature, we elaborate on a competition process of eliminating irrelevant image level predictions and a coordination process to enhance the feature across all levels. Extensive experiments are executed on two popular datasets, and our method achieves state-of-the-art performance with multi-level interpretation results. Code is available at: https://github.com/bisno/CocoER.

Reasoning-focused large language models (LLMs) are rapidly evolving across various domains, yet their capabilities in handling complex legal problems remains underexplored.In this paper, we introduce Unilaw-R1, a large language model tailored for legal reasoning.With a lightweight 7-billion parameter scale, Unilaw-R1 significantly reduces deployment cost while effectively tackling three core challenges in the legal domain: insufficient legal knowledge, unreliable reasoning logic, and weak business generalization.To address these issues, we first construct Unilaw-R1-Data, a high-quality dataset containing ∼17K distilled and screened chain-of-thought (CoT) samples.Based on this, we adopt a two-stage training strategy combining Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL), which significantly boosts the model's performance on complex legal reasoning tasks and supports interpretable decision-making in legal AI applications.To assess legal reasoning ability, we also introduce Unilaw-R1-Eval, a dedicated benchmark designed to evaluate models across single-and multi-choice legal tasks.Unilaw-R1 demonstrates strong results on authoritative benchmarks, outperforming all models of similar scale and achieving performance on par with the much larger DeepSeek-R1-Distill-Qwen-32B (54.9%).Following domainspecific training, it also showed significant gains on LawBench and LexEval, exceeding Qwen-2.5-7B-Instruct(46.6%) by an average margin of 6.6%.Code is available at: https://github.com/Hanscal/Unilaw-R1.

Analogical reasoning is fundamental to human cognition and plays a crucial role across various fields. However, previous studies have primarily focused on single-modal analogical reasoning, often ignoring the benefits of incorporating structural knowledge. Research in cognitive psychology has shown that information from multimodal sources brings more powerful cognitive transfer than single-modality sources. Motivated by the cognitive analogical theory, we address the challenge of multimodal analogical reasoning over multimodal knowledge graphs. Our approach introduces a novel triplet interaction mechanism, where both internal and external interactions between entities and relations are leveraged to enhance the accuracy of multimodal analogical reasoning. Experimental results show that our model outperforms Transformer-based multimodal pre-trained baselines. This improvement underscores the essential role of triplet interactions in integrating multimodal data, leading to more accurate analogical reasoning processes.

Emotion recognition is an important task in computer vision. However, current approaches using hard associations (e.g., element-wise addition or concatenation) suffer from information pollution. To overcome these challenges and utilize information at different scales, we present a novel Transformer-based emotion recognition (TransEmo) framework utilizing only the RGB modality as input. TransEmo is a hierarchical feature fusion framework with soft-association to effectively realize contextual interaction and emotion reasoning. It comprises a top-down path and a bottom-up path with cross-attention layers. The top-down path enriches the features of the current hierarchy gradually by using information from a larger receptive field and interacts effectively with contexts to suppress redundant information and noise. The bottom-up path introduces a set of learnable prototypes to adaptively capture clues for various emotions by reasoning with hierarchical features. Our proposed approach achieves state-of-the-art performance on two benchmarks and provides an analysis to discuss the advantages of our modules.

Generating emotion-specific talking head videos from audio input is an important and complex challenge for human-machine interaction. However, emotion is highly abstract concept with ambiguous boundaries, and it necessitates disentangled expression parameters to generate emotionally expressive talking head videos. In this work, we present EmoHead to synthesize talking head videos via semantic expression parameters. To predict expression parameter for arbitrary audio input, we apply an audio-expression module that can be specified by an emotion tag. This module aims to enhance correlation from audio input across various emotions. Furthermore, we leverage pre-trained hyperplane to refine facial movements by probing along the vertical direction. Finally, the refined expression parameters regularize neural radiance fields and facilitate the emotion-consistent generation of talking head videos. Experimental results demonstrate that semantic expression parameters lead to better reconstruction quality and controllability.

In this paper, we develop a weakly supervised learning algorithm to learn robust semantic correspondences from large-scale datasets with only image-level labels. Following the spirit of multiple instance learning (MIL), we decompose the weakly supervised correspondence learning problem into three stages: image-level matching, region-level matching, and pixel-level matching. We propose a novel cascaded online correspondence refinement algorithm to integrate MIL and the correspondence filtering and refinement procedure into a single deep network and train this network end-to-end with only image-level supervision, i.e., without point-to-point matching information. During the correspondence learning process, pixel-to-pixel matching pairs inferred from weak supervision are propagated, filtered, and enhanced through masked correspondence voting and calibration. Besides, we design a correspondence consistency check algorithm to select images with discriminative key points to generate pseudo-labels for classical matching algorithms. Finally, we filter out about 110,000 images from the ImageNet ILSVRC training set to formulate a new dataset, called SC-ImageNet. Experiments on several popular benchmarks indicate that pre-training on SC-ImageNet can improve the performance of state-ofthe-art algorithms efficiently. Our project is available on https://github.com/21210240056/SC-ImageNet.

Diffusion models (DMs) have recently gained attention with state-of-the-art performance in text-to-image synthesis. Abiding by the tradition in deep learning, DMs are trained and evaluated on the images with fixed sizes. However, users are demanding for various images with specific sizes and various aspect ratio. This paper focuses on adapting text-to-image diffusion models to handle such variety while maintaining visual fidelity. First we observe that, during the synthesis, lower resolution images suffer from incomplete object portrayal, while higher resolution images exhibit repetitively disordered presentation. Next, we establish a statistical relationship indicating that attention entropy changes with token quantity, suggesting that models aggregate spatial information in proportion to image resolution. The subsequent interpretation on our observations is that objects are incompletely depicted due to limited spatial information for low resolutions, while repetitively disorganized presentation arises from redundant spatial information for high resolutions. From this perspective, we propose a scaling factor to alleviate the change of attention entropy and mitigate the defective pattern observed. Extensive experimental results validate the efficacy of the proposed scaling factor, enabling models to achieve better visual effects, image quality, and text alignment. Notably, these improvements are achieved without additional training or fine-tuning techniques.

We introduce a scalable and fast method for solving distributionally robust optimization (DRO). Previous works have demonstrated that DRO outperforms empirical risk on a collection of inconsistent distribution of test data (the property of “uncertainty set”). However, DRO is hard to be applied for large-scale datasets and large parameterized model, due to the datapoint-level and non-differentiable objective function. In this paper, we formalize the DRO problem with the supremum of a family of subgroup-level loss functions. Subgroup loss is the cost function of partitioned uncertainty set. Then we implement the maximum of subgroup loss as the objective function and update model parameters by reweighting the descent direction, calculated from a differentiable objective function. Experimental results unveil that large parameterized models with the proposed method successfully adapt to uncertainty set whether the distribution contains out-of-domain or imbalanced property. Remarkably, with the explored reweighting strategy, the proposed algorithm effectively achieves competitive performance and robustness.