BLens uses contrastive learning to generate binary function names, achieving an F1 score of 0.79.

Tristan Benoit, Yunru Wang, Moritz Dannehl et al.

Proposed RFedAGS, a Riemannian federated learning algorithm based on gradient stream averaging, effectively handles partial participation and data heterogeneity.

Zhenwei Huang, Wen Huang, Pratik Jawanpuria et al.

CoxKAN integrates Kolmogorov-Arnold networks with symbolic regression for interpretable, high-performance survival analysis, validated on multiple datasets.

William Knottenbelt, William McGough, Rebecca Wray et al.

Proposes a unified framework combining tensor decomposition and automatic rank search, achieving high compression of pre-trained neural networks.

Ali Aghababaei-Harandi, Massih-Reza Amini

Visual prompting in MLLMs enhances visual understanding and reasoning capabilities.

Junda Wu, Zhehao Zhang, Yu Xia et al.

Reveals MDMs are essentially time-agnostic masked models, introduces efficient first-hitting sampling, questions their superiority.

Kaiwen Zheng, Yongxin Chen, Hanzi Mao et al.

ToolACE employs self-evolution synthesis of 26,507 APIs, enabling high-performance function calling with only 8B parameters, rivaling GPT-4.

Weiwen Liu, Xu Huang, Xingshan Zeng et al.

Proposes Loss-Free Balancing, a method that dynamically adjusts expert biases to improve load balance without interference gradients, outperforming auxiliary loss methods.

Lean Wang, Huazuo Gao, Chenggang Zhao et al.

Distilling large Transformers into linear RNN (Mamba) with hardware-aware decoding boosts inference speed, matching original performance.

Junxiong Wang, Daniele Paliotta, Avner May et al.

RTF framework combines hierarchical ReAct with dynamic tool retrieval, significantly improving forecasting accuracy beyond baseline models and rivaling human predictions.

Elvis Hsieh, Preston Fu, Jonathan Chen

Proposes MOHAWK: a multi-stage distillation method transforming pretrained Transformers into efficient state-space models, boosting performance with minimal data.

Aviv Bick, Kevin Y. Li, Eric P. Xing et al.

Proposes fixed and adaptive kNN-UCB methods for unbounded contexts, achieving near-minimax regret bounds, with the adaptive approach nearly optimal across all regimes.

Puning Zhao, Rongfei Fan, Shaowei Wang et al.

Proposes a derivative-free guidance method using soft value functions integrated into diffusion models for efficient goal-oriented generation without model fine-tuning.

Xiner Li, Yulai Zhao, Chenyu Wang et al.

Unsupervised multimodal embedding guides GFlowNets to generate molecules mimicking cell morphology.

Stephen Zhewen Lu, Ziqing Lu, Ehsan Hajiramezanali et al.

Open-source JumpReLU-based sparse autoencoders trained on all layers of Gemma 2, enhancing interpretability of large language models.

Tom Lieberum, Senthooran Rajamanoharan, Arthur Conmy et al.

Bloom unifies observational and interventional causal discovery through bilevel polynomial optimization with SDP-based convergence and optimality guarantees.

Qiu Chengbo, Yang Kai

Proposes a compute-optimal strategy using adaptive resource allocation, outperforming model parameter scaling with over 4x efficiency gains.

Charlie Snell, Jaehoon Lee, Kelvin Xu et al.

Repeated sampling enhances inference coverage; on SWE-bench Lite, coverage increased from 15.9% to 56% with 250 samples, outperforming single-sample SOTA (43%).

Bradley Brown, Jordan Juravsky, Ryan Ehrlich et al.

Informed corrector using model guidance and hollow Transformer architecture improves discrete diffusion sampling efficiency and quality.

Yixiu Zhao, Jiaxin Shi, Feng Chen et al.

Proposes a Stackelberg game-based robust meta-learning framework using normalizing flows for explicit task distribution modeling, enhancing adaptation under distribution shifts.

Cheems Wang, Yiqin Lv, Yixiu Mao et al.