63 accepted papers
LLMs achieve remarkable multi-step reasoning capabilities, yet effectively transferring these skills via post-training distillation remains challenging. Existing data selection methods, ranging from manual curation to heuristics based on length, entropy, or overall loss, fail to capture the causal i…
Reinforcement learning (RL) has demonstrated considerable potential for enhancing reasoning in large language models (LLMs). However, existing methods suffer from Gradient Starvation and Policy Degradation when training directly on samples with mixed difficulty. To mitigate this, prior approaches l
Reinforcement Learning with Verifiable Rewards (RLVR) is increasingly viewed as a tree pruning mechanism. However, we identify a systemic pathology termed Recursive Space Contraction (RSC), an irreversible collapse driven by the combined dynamics of positive sharpening and negative squeezing, where …
LLM-as-a-Judge has been widely adopted across various research and practical applications, yet the robustness and reliability of its evaluation remain a critical issue. A core challenge it faces is bias, which has primarily been studied in terms of known biases and their impact on evaluation outcome…
Most state-of-the-art time series imputation methods can leverage textual information to improve imputation quality, but they often struggle because they fail to effectively filter noisy information from large language model (LLM) derived textual information. Some existing solutions only filter over
Reinforcement learning (RL) has demonstrated potential for enhancing reasoning in large language models (LLMs). However, effective RL training, which requires medium-difficulty training samples, faces two fundamental challenges: Effective Data Scarcity and Dynamic Difficulty Shifts, where medium-dif…
Text-to-image (T2I) models have achieved remarkable success in generating high-fidelity images, but they often fail in handling complex spatial relationships, e.g., spatial perception, reasoning, or interaction. These critical aspects are largely overlooked by current benchmarks due to their short o…
The deployment of Large Language Models (LLMs) faces a critical bottleneck when handling lengthy inputs: the prohibitive memory footprint of the Key Value (KV) cache. To address this bottleneck, the token pruning paradigm leverages attention sparsity to selectively retain a small, critical subset of…
Reinforcement Learning (RL) can directly enhance the reasoning capabilities of large language models without extensive reliance on Supervised Fine-Tuning (SFT). In this work, we revisit the traditional Policy Gradient (PG) mechanism and propose a minimalist RL approach termed Group Policy Gradient (…
Unified Multimodal Models (UMMs) integrate both visual understanding and generation within a single framework. Their ultimate aspiration is to create a cycle where understanding and generation mutually reinforce each other. While recent post-training methods have successfully leveraged understanding…
Reinforcement Learning with Verifiable Rewards (RLVR) offers a robust mechanism for enhancing mathematical reasoning in large models. However, we identify a systematic lack of emphasis on more challenging questions in existing methods from both algorithmic and data perspectives, despite their import…
Reinforcement learning often produces high-frequency oscillatory control signals that undermine the safety and stability required for physical deployment. Explicit action chunking addresses this by predicting fixed-horizon trajectories but increases the policy output dimension to R^hd, leading to op…
A/B testing on platforms often faces challenges from network interference, where a unit's outcome depends not only on its own treatment but also on the treatments of its network neighbors. To address this, cluster-level randomization has become standard, enabling the use of network-aware estimators.…
Spatiotemporal predictive learning (STPL) aims to forecast future frames from past observations and is essential across a wide range of applications. Compared with recurrent or hybrid architectures, pure convolutional models offer superior efficiency and full parallelism, yet their fixed receptive f
Hate Video Detection (HVD) is crucial for online ecosystems. Existing methods assume identical distributions between training (source) and inference (target) data. However, hateful content often evolves into irregular and ambiguous forms to evade censorship, resulting in substantial semantic drift a
Recent advances in reinforcement learning (RL) have significantly enhanced the agentic capabilities of large language models (LLMs). In long-term and multi-turn agent tasks, existing approaches driven solely by outcome rewards often suffer from the problem of sparse supervision. To address the chall…
Large Multimodal Models (LMMs) have shown promising in-context learning (ICL) capabilities, but scaling to many-shot settings remains difficult due to limited context length and high inference cost. To address these challenges, task-vector-based methods have been explored by inserting compact repres
Millimeter-wave (mmWave) radar enables contactless respiratory sensing,yet fine-grained monitoring is often degraded by nonstationary interference from body micromotions.To achieve micromotion interference removal,we propose mmWave-Diffusion,an observation-anchored conditional diffusion framework th…
It has been demonstrated that incorporating external information as textual modality can effectively improve time series forecasting accuracy. However, current multi-modal models ignore the dynamic and different relations between time series patterns and textual features, which leads to poor perform
Pulmonary segments parcellation is essential to thoracoscopic segmentectomy. Surgeons manually outline pulmonary segments from preoperative images before surgery, which is a time-consuming, labor-intensive and mental-stress procedure. This work proposes a novel small learning model of deep support v…
Large language models (LLMs) exhibit remarkable text-generation capabilities, yet struggle with factual consistency, motivating growing interest in factuality verification. Existing factuality verification methods typically follow a Decompose-Then-Verify paradigm, which improves granularity but suff
In recent years, Transformer-based models have been widely used in time series forecasting tasks, demonstrating exceptional performance. However, these models lack interpretability, making it difficult to identify which components play a core role in predictions and which are redundant. To address t…
Self-taught reasoners (STaRs) enhance the mathematical reasoning abilities of large language models (LLMs) by leveraging self-generated responses for self-training. Recent studies have incorporated reward models to guide response selection or decoding, aiming to obtain higher-quality data. However,
Based on pseudo-labels, voxel-wise contrastive learning (VCL) is a prominent approach designed to learn effective feature representations for semi-supervised medical image segmentation. However, in multi-organ segmentation (MoS), the complex anatomical structures of certain organs often lead to many…
Visual Emotion Recognition (VER) is a critical yet challenging task aimed at inferring emotional states of individuals based on visual cues. However, existing works focus on single domains, e.g., realistic images or stickers, limiting VER models' cross-domain generalizability. To fill this gap, we…
Multimodal end-to-end autonomous driving has shown promising advancements in recent work. By embedding more modalities into end-to-end networks, the system's understanding of both static and dynamic aspects of the driving environment is enhanced, thereby improving the safety of autonomous driving. I
Positional bias (PB), manifesting as non-uniform sensitivity across different contextual locations, significantly impairs long-context comprehension and processing capabilities. Previous studies have addressed PB either by modifying the underlying architectures or by employing extensive contextual a
Unmanned Aerial Vehicles (UAVs) are increasingly utilized across commercial, military, and public safety domains, where target tracking is crucial for imagery content analysis and various applications. This letter explores the challenges in UAV target tracking, particularly addressing deficiencies i
Human Activity Recognition (HAR) such as fall detection has become increasingly critical due to the aging population, necessitating effective monitoring systems to prevent serious injuries and fatalities associated with falls. This study focuses on fine-tuning the Vision Transformer (ViT) model spec…
Cataracts are the leading cause of blindness worldwide, with surgery as the only effective treatment. Accurate prediction of Best Corrected Visual Acuity (BCVA) is crucial for surgical planning. In this paper, we propose a novel Unified Multimodal Learning (UML) framework for BCVA prediction with un…
Recent studies have highlighted the interplay between diffusion models and representation learning. Intermediate representations from diffusion models can be leveraged for downstream visual tasks, while self-supervised vision models can enhance the convergence and generation quality of diffusion mod…
Point cloud registration is a crucial task in computer vision and 3D reconstruction, aiming to align multiple point clouds to achieve globally consistent geometric structures. However, traditional point cloud registration methods face challenges when dealing with low overlap and large-scale point cl
Data heterogeneity is one of the key challenges in federated learning, and many efforts have been devoted to tackling this problem. However, distributed concept drift with data heterogeneity, where clients may additionally experience different concept drifts, is a largely unexplored area. In this wo…
In this letter, we present a framework for low frame rate tracking from unmanned aerial vehicles (UAV). This is a challenging task in which the target has to be localized in a large search space. This problem is amplified by the fact that the tracked object is most often composed of a limited number
Previous probabilistic models for 3D Human Pose Estimation (3DHPE) aimed to enhance pose accuracy by generating multiple hypotheses. However, most of the hypotheses generated deviate substantially from the true pose. Compared to deterministic models, the excessive uncertainty in probabilistic models…
The long-term consistent visual tracking of large-scale moving swarms of animals or autonomous moving robots (AMR) is extremely challenging when the three factors are involved: 1) similar appearance of animals or AMR, 2) frequent and unpredictable occlusions, and 3) non-linear maneuvers. When facing
Accurate IP geolocation is crucial for location-aware applications. While recent advances in router-centric IP graph methods have garnered attention, they face two persistent challenges: (1) the sparsity problem of IP graphs in rural areas and (2) the limited explainability of current IP geolocation…
We propose FaceCom a method for 3D facial shape completion which delivers high-fidelity results for incomplete facial inputs of arbitrary forms. Unlike end-to-end shape completion methods based on point clouds or voxels our approach relies on a mesh-based generative network that is easy to optimize…
"In recent years, 2D human pose estimation has made significant progress on public benchmarks. However, many of these approaches face challenges of less applicability in the industrial community due to the large number of parametric quantities and computational overhead. Efficient human pose estimat…
Accurate IP geolocation is indispensable for location-aware applications. While recent advances based on router-centric IP graphs are considered cutting-edge, one challenge remain: the prevalence of sparse IP graphs (14.24% with fewer than 10 nodes, 9.73% isolated) limits graph learning. To mitigate…
Pulmonary fissures are anatomical biomarkers used to evaluate the severity of chronic obstructive pulmonary disease. The completeness of the fissures is significantly associated with this disease. This work proposes a new fully automated fissure completeness assessment framework on the basis of deep…
Concept Bottleneck Models (CBMs), which break down the reasoning process into the input-to-concept mapping and the concept-to-label prediction, have garnered significant attention due to their remarkable interpretability achieved by the interpretable concept bottleneck. However, despite the transpar…
Swin-Transformer has demonstrated remarkable success in computer vision by leveraging its hierarchical feature representation based on Transformer. In speech signals, emotional information is distributed across different scales of speech features, e. g., word, phrase, and utterance. Drawing above in…
Recently model merging techniques have surfaced as a solution to combine multiple single-talent models into a single multi-talent model. However previous endeavors in this field have either necessitated additional training or fine-tuning processes or require that the models possess the same pre-trai…
Gait recognition is widely used in diversified practical applications. Currently, the most prevalent approach is to recognize human gait from RGB images, owing to the progress of computer vision technologies. Nevertheless, the perception capability of RGB cameras deteriorates in rough circumstances,…
In this work, we propose an ID-preserving talking head generation framework, which advances previous methods in two aspects. First, as opposed to interpolating from sparse flow, we claim that dense landmarks are crucial to achieving accurate geometry-aware flow fields. Second, inspired by face-swapp…
Online A/B tests have become increasingly popular and important for social platforms. However, accurately estimating the global average treatment effect (GATE) has proven to be challenging due to network interference, which violates the Stable Unit Treatment Value Assumption (SUTVA) and poses great…
Temporal knowledge graph (TKG) has been proved to be an effective way for modeling dynamic facts in real world. Many efforts have been devoted into predicting future events i.e. extrapolation, on TKGs. Recently, rule-based knowledge graph completion methods which are considered to be more interpreta…
Multi-intent natural language understanding (NLU) has recently gained attention. It detects multiple intents in an utterance, which is better suited to real-world scenarios. However, the state-of-the-art joint NLU models mainly detect multiple intents on threshold-based strategy, resulting in one ma…
Non-autoregressive translation (NAT) models, which eliminate the sequential dependencies within the target sentence, have achieved remarkable inference speed, but suffer from inferior translation quality. Towards exploring the underlying causes, we carry out a thorough preliminary study on the atten…
Despite the tantalizing success in a broad of vision tasks, transformers have not yet demonstrated on-par ability as ConvNets in high-resolution image generative modeling. In this paper, we seek to explore using pure transformers to build a generative adversarial network for high-resolution image sy…
Pre-training language models have achieved thriving success in numerous natural language understanding and autoregressive generation tasks, but non-autoregressive generation in applications such as machine translation has not sufficiently benefited from the pre-training paradigm. In this work, we es…
Human motion prediction is essential in human-robot interaction. Current research mostly considers the joint dependencies but ignores the bone dependencies and their relationship in the human skeleton, thus limiting the prediction accuracy. To address this issue, we represent the human skeleton as a…
Recently, many model quantization approaches have been investigated to reduce the model size and improve the inference speed of convolutional neural networks (CNNs). However, these approaches usually inevitably lead to a decrease in classification accuracy. To address this problem, this paper propos…
Self-training is a competitive approach in domain adaptive segmentation, which trains the network with the pseudo labels on the target domain. However inevitably, the pseudo labels are noisy and the target features are dispersed due to the discrepancy between source and target domains. In this paper…
Leveraging lexical constraint is extremely significant in domain-specific machine translation and interactive machine translation. Previous studies mainly focus on extending beam search algorithm or augmenting the training corpus by replacing source phrases with the corresponding target translation.…
Nowadays, speech spoofing is so common that it presents a great challenge to social security. Thus, it is of great significance to recognize a spoofed speech from a genuine one. Most of the current researches have focused on voice conversion (VC), synthesis and recapture which mimic a target speaker…
In this paper, we propose a novel pose optimizer which can be inserted into either supervised or unsupervised end-to-end visual odometry for the purpose of local pose optimization. The pose optimizer is an analogue of the pose graph optimization used in traditional VSLAM algorithms. Local pose optim…
As the rapid development of precision medicine, in-vivo manipulation of micro/nano-scaled particles has attracted increasing attention in recent years. The collision is one of the main reasons that falls the in-vivo particle transportation fail. In this paper, we develop an in-vivo cell transportati…
As rapid development of precision medicine, in vivo manipulation of micro/nano-scaled particles have attracted increasing attention in recent years. To accommodate complex in-vivo environment, robot-aided automated manipulation technology is highly demanded in trapping and controlling micro/nano-par…
We propose an online tracking algorithm in which the object tracking is achieved by using subspace learning and non-negative matrix factorization (NMF) under the partile filtering framework. The object appearance is modeled by a non-negative combination of non-negative components learned from exampl…
In this paper, we cast tracking as a novel multi-task learning problem and exploit various types of visual features. We use an on-line feature selection mechanism based on the two-class variance ratio measure, applied to log likelihood distributions computed with respect to a given feature from samp…
Optical tweezers has become a powerful tool in automated cell transportation control and has been used in a variety of biological applications. The use of optical tweezers for cell surgery has great potential for various biomedical applications such as microinjection, organelle extraction and modifi…