π― What it does: This paper proposes WAH-MVC, a multi-view clustering method that employs Wasserstein-aligned hyperbolic encoding on the Lorentz manifold.
CodeObject DetectionSegmentationConvolutional Neural NetworkImagePhysics Related
π― What it does: A physics-driven WavePCNet network is studied for detecting and segmenting occluded objects in non-line-of-sight (NLOS) scenes using sparse light spot images without active illumination.
π― What it does: Achieve 3D editing of controllable intensity multi-weather (rain, snow, fog) from ordinary scenes by fusing multi-weather style diffusion models and combining with 4D Gaussian fields;
π― What it does: Proposed the WeightFlow framework, which directly captures the probability density evolution of stochastic dynamics by modeling probability distributions in the neural network weight space and learning the continuous evolution of weight graphs;
WenetSpeech-Yue: A Large-Scale Cantonese Speech Corpus with Multi-dimensional Annotation
Longhao Li (Northwestern Polytechnical University), Lei Xie (Hong Kong University of Science and Technology)
CodeData SynthesisTransformerLarge Language ModelSupervised Fine-TuningMultimodalityBenchmarkAudio
π― What it does: Built the WenetSpeech-Pipe data processing pipeline, and used this pipeline to collect and annotate 21,800 hours of Cantonese speech data, generating the WenetSpeech-Yue large-scale corpus, while releasing the WSYue-eval benchmark set covering ASR and TTS evaluations.
What Makes a Good Speech Tokenizer for LLM-Centric Speech Generation? A Systematic Study
Xiaoran Fan (Fudan University), Tao Gui (Fudan University)
CodeGenerationTransformerLarge Language ModelSupervised Fine-TuningTextAudio
π― What it does: Systematically investigates the impact of speech tokenizer design in LLM-centric speech generation on cross-modal alignment and speech quality, and proposes a multi-word prediction (MTP) and speaker-aware generation scheme.
What to Ask Next? Probing the Imaginative Reasoning of LLMs with TurtleSoup Puzzles
Mengtao Zhou (Huazhong University of Science and Technology), Bang Liu (Huazhong University of Science and Technology)
CodeLarge Language ModelTextBenchmarkChain-of-Thought
π― What it does: This study proposes an interactive benchmark called TurtleSoup-Bench centered on turtle soup riddles, and constructs a multi-stage Mosaic-Agent model to evaluate the imaginative reasoning capabilities of large language models in information-scarce environments.
When Eyes and Ears Disagree: Can MLLMs Discern Audio-Visual Confusion?
Qilang Ye (Nankai University), Yu Zhou (Nankai University)
CodeTransformerLarge Language ModelReinforcement LearningVision Language ModelVideoMultimodalityBenchmarkAudio
π― What it does: Proposed the AVConfuseBench audio-visual confusion benchmark and designed the RL-CoMM method to enhance the reasoning and answer accuracy of multi-modal large language models in scenarios where audio is missing or tampered with.
When Genes Speak: A Semantic-Guided Framework for Spatially Resolved Transcriptomics Data Clustering
Jiangkai Long (China University of Geosciences), Xuesong Yan (China University of Geosciences)
CodeClassificationGraph Neural NetworkLarge Language ModelBiomedical Data
π― What it does: Propose the SemST framework, integrating the semantic embeddings of gene symbols with spatial graph neural networks to achieve clustering of spatial transcriptomic data.
When Natural Strategies Meet Fuzziness and Resource-Bounded Actions
Marco Aruta (University of Naples Federico II), Aniello Murano (University of Naples Federico II)
CodeExplainability and InterpretabilityReinforcement LearningTabular
π― What it does: This paper proposes the HumanATL[F] logic, combining natural strategies with fuzzy semantics and consumable resource constraints to construct interpretable and budget-constrained multi-agent strategies.
When Smiley Turns Hostile: Interpreting How Emojis Trigger LLMsβ Toxicity
Shiyao Cui (Tsinghua University), Minlie Huang (Tsinghua University)
CodeSafty and PrivacyExplainability and InterpretabilityLarge Language ModelPrompt EngineeringText
π― What it does: This paper constructs emoji-containing prompts (rewriting AdvBench and adversarial prompts) to systematically evaluate the effectiveness of emojis in triggering toxic generation in LLMs across multi-lingual, multi-model, and multi-attack scenarios.
π― What it does: Propose a task vector insertion framework STV based on perceptual sensitivity to achieve multi-sample in-context learning for multimodal large models without increasing context length.
CodeData SynthesisTransformerLarge Language ModelTextBenchmark
π― What it does: This paper proposes a diagnostic test set named SNIC to evaluate the reasoning capabilities of large language models (LLMs) in social norm-driven reference resolution; it is programmatically expanded from 51 human-validated scenarios to 9,000 instances, focusing on norms in physical environments (e.g., cleanliness, service, and use of clean items).
π― What it does: Propose the WhisperDiari framework, achieving simultaneous speaker separation and ASR in the token space, supporting synchronized generation of speaker labels and text.
Whispering Agents: A Event-Driven Covert Communication Protocol for the Internet of Agents
Kaibo Huang (Beijing University of Posts and Telecommunications), Linna Zhou (Beijing University of Posts and Telecommunications)
CodeSafty and PrivacyLarge Language ModelText
π― What it does: Proposed an event-driven covert communication protocol called Pi-CCAP for the Internet of Agents (IoA), and presented a unified covert event channel model (storage, timing, and behavior dimensions);
Who Is Helping Whom? Analyzing Inter-Dependencies to Evaluate Cooperation in Human-AI Teaming
Upasana Biswas (Arizona State University), Subbarao Kambhampati (Arizona State University)
CodeReinforcement LearningAgentic AI
π― What it does: This work proposes a constructive mutual dependency metric to evaluate the collaboration of human-robot teams, and conducts user experiments with SOTA zero-shot cooperative agents in the Overcooked environment.
Who Should I Trust? Explicit Confidence-Focused Multimodal Intent Recognition
Yi Liu (Xinjiang University), Lanlan Lu (Xinjiang University)
CodeRecognitionExplainability and InterpretabilityTransformerVideoTextPoint CloudAudio
π― What it does: This paper proposes an explicit confidence attention-based multimodal intent recognition framework called ECFMIR, which uses CLens to estimate confidence for each modality and cross-modal features and then weight them for fusion.
CodeData SynthesisAI Code AssistantTransformerLarge Language ModelSupervised Fine-TuningPrompt EngineeringTabularBenchmarkChain-of-Thought
π― What it does: This paper systematically evaluates the three core capabilities of open-source LLMs in data analysis tasks and proposes a data synthesis framework based on moderate-length dialogues, moderately difficult instances, and concise reasoning summaries to significantly enhance the model's analytical reasoning performance.
Wi-CBR: Salient-aware Adaptive WiFi Sensing for Cross-domain Behavior Recognition
Ruobei Zhang (Hefei University Of Technology), Jiabao Guo (Guizhou Normal University)
CodeRecognitionDomain AdaptationConvolutional Neural NetworkTransformerContrastive LearningTime Series
π― What it does: Proposed the Wi-CBR framework, which fuses WiFi phase and DFS signals through dual-branch self-attention and salience-guided modules to achieve cross-domain behavior recognition.
WorldRFT: Latent World Model Planning with Reinforcement Fine-Tuning for Autonomous Driving
Pengxuan Yang (State Key Laboratory Of Multimodal Artificial Intelligence Systems Institute Of Automation Chinese Academy Of Sciences), Qichao Zhang (Li Auto)
π― What it does: Proposes WorldRFT, a planning-oriented potential world model framework that integrates spatial perception encoding, hierarchical planning refinement, and reinforcement learning fine-tuning to enhance the safety and accuracy of end-to-end autonomous driving.
WRitEer: A Multi-Objective, Preference-Driven Multi-Agent Framework for Human-Like Advanced Text Generation
Junchuan Yu (Tianjin University), Yuyang Sun (Huazhong Agricultural University)
CodeGenerationOptimizationLarge Language ModelReinforcement LearningAgentic AIPrompt EngineeringTextBenchmark
π― What it does: Proposed the WRitEer framework, enabling interactive iterative writing through three agents (Reader, Editor, Writer), enhancing text naturalness and emotional expression.
X-ReID: Multi-granularity Information Interaction for Video-Based Visible-Infrared Person Re-Identification
Chenyang Yu, Huchuan Lu (Dalian University Of Technology)
CodeRetrievalTransformerVision Language ModelContrastive LearningVideoMultimodality
π― What it does: Study visible-infrared person re-identification in videos, propose the X-ReID framework, and achieve new state-of-the-art results on two public datasets.
Hao Wang (Sun Yat-sen University), Xiaodan Liang (Sun Yat-sen University)
CodeSegmentationTransformerLarge Language ModelSupervised Fine-TuningMultimodality
π― What it does: Propose the X-SAM framework, extending the Segment Anything Model into a unified 'from segment anything to any segmentation' model that supports dual queries of text and vision.
X2Edit: Revisiting Arbitrary-Instruction Image Editing Through Self-Constructed Data and Task-Aware Representation Learning
Jian Ma (OPPO AI Center), Haonan Lu (OPPO AI Center)
CodeGenerationTransformerMixture of ExpertsVision Language ModelDiffusion modelContrastive LearningImageMultimodalityBenchmark
π― What it does: Constructed the X2Edit dataset, which scales up to 3.7M and covers 14 editing tasks, and developed a lightweight, plug-and-play arbitrary instruction image editing model X2Edit based on FLUX.1.
π― What it does: Propose a lightweight MLP model called XLinear, which leverages learnable global tokens and time/variable gating mechanisms to integrate endogenous sequences with exogenous drivers for long-term forecasting.
π― What it does: Propose a training-free cross-modal LiDAR semantic segmentation framework named xMHashSeg, which leverages a base model and non-parametric network to extract features from 2D images, depth maps, and 3D point clouds, and achieves unlabeled, no-additional-training point cloud semantic segmentation through hash learning for cross-modal feature fusion.
π― What it does: Proposes the LogicRAG framework, which dynamically constructs a query-dependent directed acyclic graph (DAG) during inference and performs adaptive retrieval, eliminating the need for pre-built graphs.
π― What it does: This paper investigates the issue of false image misjudgment caused by distribution drift in AI-generated image detection, proposing a post-scalar calibration method to dynamically adjust the decision threshold and restore detection capabilities for new generative models.
Zero-Reference Joint Low-Light Enhancement and Deblurring via Visual Autoregressive Modeling with VLM-Derived Modulation
Wei Dong (McMaster University), Jun Chen (McMaster University)
CodeRestorationVision Language ModelImage
π― What it does: Proposed a fully unsupervised generative framework called VAR-LIDE, which combines a visual autoregressive model with a visual language model (VLM) prior to achieve joint recovery of low-light enhancement and deblurring.
π― What it does: Proposed a zero-shot unsupervised implicit neural manifold representation (INMR) to achieve dynamic MRI reconstruction with extremely high spatiotemporal resolution.
Zero-Shot Open-Vocabulary Human Motion Grounding with Test-Time Training
Yunjiao Zhou (Nanyang Technological University), Jianfei Yang (Nanyang Technological University)
CodeSegmentationRetrievalLarge Language ModelVision-Language-Action ModelVideoText
π― What it does: Propose a zero-shot, annotation-free open-vocabulary human action segmentation framework called ZOMG, which leverages a large language model to split text sub-actions and achieves instance-level temporal segmentation on a pre-trained encoder through soft mask optimization.
π― What it does: Propose a zero-shot robotic manipulation framework RobMRAG based on multi-modal retrieval-augmented generation (MRAG), which leverages multi-source knowledge bases for retrieval and achieves precise pose alignment through 3D Gaussian Splatting.
π― What it does: Propose a two-stage reference-based video appearance editing framework called Zero-to-Hero: first, edit the anchor frame into a reference image using a zero-shot approach, then propagate the appearance consistently to all frames, and in the second stage, use a conditional generation model to restore distortions caused by the zero-shot propagation.
π― What it does: Proposed a large-scale local forgery image dataset named BR-Gen, and designed a noise-guided amplification attention visual Transformer named NFA-ViT for detecting and locating fine-grained forgeries.