Proposed Stochastic Concept Bottleneck Model (SCBM) models concept dependencies via multivariate normal distribution, significantly improving intervention effectiveness on synthetic and real datasets.

Moritz Vandenhirtz, Sonia Laguna, Ričards Marcinkevičs et al.

Proposed STBench benchmark evaluates 13 LLMs across four spatio-temporal abilities, emphasizing knowledge comprehension and reasoning, with over 60,000 QA pairs.

Wenbin Li, Di Yao, Ruibo Zhao et al.

Proposes Dynamic Gaussian Marbles for monocular video view synthesis, combining isotropic Gaussian modeling, hierarchical merging, and priors to improve scene reconstruction.

Colton Stearns, Adam Harley, Mikaela Uy et al.

RouteLLM trains preference-based routers, reducing costs over 2x while maintaining response quality.

Isaac Ong, Amjad Almahairi, Vincent Wu et al.

This paper surveys Mixture of Experts in LLMs, proposes a new taxonomy, and analyzes its applications and future directions.

Weilin Cai, Juyong Jiang, Fan Wang et al.

This paper systematically evaluates the current state of membership inference attacks (MIAs) on LLMs, highlighting dataset distribution shifts and proposing multiple mitigation strategies.

Matthieu Meeus, Igor Shilov, Shubham Jain et al.

MotionBooth enables motion-aware text-to-video generation using few images, enhancing video quality.

Jianzong Wu, Xiangtai Li, Yanhong Zeng et al.

Univariate Skeleton Prediction using Transformers excels in multivariate systems, surpassing existing methods.

Giorgio Morales, John W. Sheppard

Proposes MedMNIST-C benchmark with targeted corruptions to improve medical image classifier robustness, leveraging domain knowledge for efficient data augmentation.

Francesco Di Salvo, Sebastian Doerrich, Christian Ledig

DriftLens is an unsupervised real-time concept drift detection framework with detection accuracy up to 85%.

Salvatore Greco, Bartolomeo Vacchetti, Daniele Apiletti et al.

GraphPipe introduces graph-based pipeline parallelism, boosting large-scale DNN training throughput by up to 1.6× and reducing search time by 9-21×.

Byungsoo Jeon, Mengdi Wu, Shiyi Cao et al.

StableNormal reduces diffusion inference variance to produce stable, sharp surface normals without ensembling, excelling in complex scenes.

Chongjie Ye, Lingteng Qiu, Xiaodong Gu et al.

Dreamitate fine-tunes a video diffusion model for robot visuomotor tasks, outperforming behavior cloning in generalization, with success rates over 85% in four tasks.

Junbang Liang, Ruoshi Liu, Ege Ozguroglu et al.

Cambrian-1 introduces a vision-centric multimodal large language model (MLLM) leveraging 20+ vision encoders and a novel Spatial Vision Aggregator (SVA), achieving state-of-the-art performance.

Shengbang Tong, Ellis Brown, Penghao Wu et al.

Introduces Ragnarök, a reusable end-to-end RAG framework, with MS MARCO V2.1, establishing the TREC 2024 RAG track; baselines include GPT-4o and Command R+.

Ronak Pradeep, Nandan Thakur, Sahel Sharifymoghaddam et al.

Unified framework linking reward models, parameter updates, and prompts via six bidirectional transformations.

Deng Cai, Huayang Li, Tingchen Fu et al.

DemoRank improves demonstration selection for LLMs in ranking tasks using dependency-aware reranking, significantly boosting performance.

Wenhan Liu, Yutao Zhu, Zhicheng Dou

This study evaluates LLMs (GPT-4, Gemini, Claude) in NLP peer review, revealing high defect rates and limited professional judgment, with scores averaging 7.45 vs. human 6.41.

Jiangshu Du, Yibo Wang, Wenting Zhao et al.

Proposes Trace framework converting complex workflows into OPTO problems, leveraging execution traces and LLMs for automatic parameter optimization, outperforming specialized optimizers.

Ching-An Cheng, Allen Nie, Adith Swaminathan

Logic-LM++ enhances symbolic reasoning accuracy through multi-step refinement, achieving an average improvement of 18.5%.

Shashank Kirtania, Priyanshu Gupta, Arjun Radhakirshna