Suchergebnisse - (coding OR Decoding) (algorithm OR algorithmsssic)

Language Models that Think, Chat Better von Bhaskar, Adithya, Ye, Xi, Chen, Danqi

Inhaltsangabe: “… rule-based rewards in verifiable domains such as mathematics and code. However, RLVR leads to limited …”
Volltext

Active Attacks: Red-teaming LLMs via Adaptive Environments von Yun, Taeyoung, St-Charles, Pierre-Luc, Park, Jinkyoo, Bengio, Yoshua, Kim, Minsu

Inhaltsangabe: “… exploration, we introduce \textit{Active Attacks}, a novel RL-based red-teaming algorithm that adapts its …”
Volltext

From Static to Dynamic: Adaptive Monte Carlo Search for Mathematical Process Supervision von Ma, Jie, Qi, Shihao, Xing, Rui, Yin, Ziang, Wei, Bifan, Liu, Jun, Liu, Tongliang

Inhaltsangabe: “… , it enhances the path expansion through a Monte Carlo algorithm with a temporally adaptive policy that begins …”
Volltext

Vision-Zero: Scalable VLM Self-Improvement via Strategic Gamified Self-Play von Wang, Qinsi, Liu, Bo, Zhou, Tianyi, Shi, Jing, Lin, Yueqian, Chen, Yiran, Li, Hai Helen, Wan, Kun, Zhao, Wentian

Inhaltsangabe: “… -SPO), a novel training algorithm that alternates between Self-Play and reinforcement learning …”
Volltext

Who Should Have a Place on the Ark? Parameterized Algorithms for the Maximization of Phylogenetic Diversity von Schestag, Jannik

Inhaltsangabe: “… after which they can no longer be saved.For Time-PD,we present color-coding algorithms that prove …”
Volltext

FOLK: Fast Open-Vocabulary 3D Instance Segmentation via Label-guided Knowledge Distillation von Wu, Hongrui, Gao, Zhicheng, Cao, Jin, Yao, Kelu, Shen, Wen, Wei, Zhihua

Inhaltsangabe: “… propose a label-guided distillation algorithm to distill open-vocabulary knowledge from label-consistent 2 …”
Volltext

Probabilistic Hyper-Graphs using Multiple Randomly Masked Autoencoders for Semi-supervised Multi-modal Multi-task Learning von Mihai-Cristian, Pîrvu, Leordeanu, Marius

Inhaltsangabe: “… the distribution of hyper-edges on each forward pass. Additionally, the model adapts the standard MAE algorithm …”
Volltext

FedAdamW: A Communication-Efficient Optimizer with Convergence and Generalization Guarantees for Federated Large Models von Liu, Junkang, Shang, Fanhua, Liu, Hongying, Tian, Yuxuan, Liu, Yuanyuan, Liu, Jin, Zhu, Kewen, Lin, Zhouchen

Inhaltsangabe: “… \underline{AdamW} algorithm, called \texttt{FedAdamW}, for training and fine-tuning various large models …”
Volltext

CORGI: Efficient Pattern Matching With Quadratic Guarantees von Weitekamp, Daniel

Inhaltsangabe: “… automatically generate rules from example-driven induction or code synthesis, they can easily produce worst-case …”
Volltext

MoDES: Accelerating Mixture-of-Experts Multimodal Large Language Models via Dynamic Expert Skipping von Huang, Yushi, Wang, Zining, Yuan, Zhihang, Ding, Yifu, Gong, Ruihao, Guo, Jinyang, Liu, Xianglong, Zhang, Jun

Inhaltsangabe: “… the optimal thresholds, we introduce a frontier search algorithm that exploits monotonicity properties …”
Volltext

Solving Spatial Supersensing Without Spatial Supersensing von Udandarao, Vishaal, Karthik, Shyamgopal, Nath, Surabhi S., Hochlehnert, Andreas, Bethge, Matthias, Prabhu, Ameya

Inhaltsangabe: “… object-count predictions unchanged; instead, Cambrian-S inference algorithm relies largely on a shortcut …”
Volltext

Spira: Exploiting Voxel Data Structural Properties for Efficient Sparse Convolution in Point Cloud Networks von Adamopoulos, Dionysios, Poulopoulou, Anastasia, Goumas, Georgios, Giannoula, Christina

Inhaltsangabe: “… for GPUs. Spira proposes (i) a high-performance one-shot search algorithm that builds the kernel map …”
Volltext

MICCAI STS 2024 Challenge: Semi-Supervised Instance-Level Tooth Segmentation in Panoramic X-ray and CBCT Images von Wang, Yaqi, Li, Zhi, Wu, Chengyu, Liu, Jun, Zhang, Yifan, Ni, Jiaxue, Luo, Qian, Chen, Jialuo, Zhang, Hongyuan, Liu, Jin, Han, Can, Fu, Kaiwen, Ji, Changkai, Cai, Xinxu, Hao, Jing, Zheng, Zhihao, Xu, Shi, Chen, Junqiang, Zhang, Qianni, Qian, Dahong, Wang, Shuai, Zhou, Huiyu

Inhaltsangabe: “… ) registered teams. To ensure algorithmic excellence and full transparency, we rigorously evaluated the valid …”
Volltext

Semantic Soft Bootstrapping: Long Context Reasoning in LLMs without Reinforcement Learning von Mitra, Purbesh, Ulukus, Sennur

Inhaltsangabe: “… optimization (GRPO), which is a commonly used RLVR algorithm. Our code is available at https://github.com …”
Volltext

How Far are Modern Trackers from UAV-Anti-UAV? A Million-Scale Benchmark and New Baseline von Zhang, Chunhui, Liu, Li, Zhang, Zhipeng, Wang, Yong, Wen, Hao, Zhou, Xi, Ge, Shiming, Wang, Yanfeng

Inhaltsangabe: “… of 50 modern deep tracking algorithms demonstrates that there is still significant room for improvement …”
Volltext

TraPO: A Semi-Supervised Reinforcement Learning Framework for Boosting LLM Reasoning von Yang, Shenzhi, Zhu, Guangcheng, Zheng, Xing, MA, Yingfan, Chen, Zhongqi, Song, Bowen, Wang, Weiqiang, Zhao, Junbo, Chen, Gang, Wang, Haobo

Inhaltsangabe: “… an effective policy optimization algorithm, TraPO, that identifies reliable unlabeled samples by matching …”
Volltext

HERO-Sign: Hierarchical Tuning and Efficient Compiler-Time GPU Optimizations for SPHINCS+ Signature Generation von Zhou, Yaoyun, Wang, Qian

Inhaltsangabe: “… algorithm that adapts fusion schemes to different GPU architectures. To further improve performance, HERO …”
Volltext

SP-Rank: A Dataset for Ranked Preferences with Secondary Information von Hosseini, Hadi, Mandal, Debmalya, Puhan, Amrit

Inhaltsangabe: “… algorithms that leverage both first-order preferences and second-order predictions in ranking tasks. Each …”
Volltext

TowerMind: A Tower Defence Game Learning Environment and Benchmark for LLM as Agents von Wang, Dawei, Zhou, Chengming, Zhao, Di, Liu, Xinyuan, Ma, Marci Chi, Ushaw, Gary, Davison, Richard

Inhaltsangabe: “… evaluate two classic reinforcement learning algorithms: Ape-X DQN and PPO. By offering a lightweight …”
Volltext

Streaming Operator Inference for Model Reduction of Large-Scale Dynamical Systems von Koike, Tomoki, Mohan, Prakash, de Frahan, Marc T. Henry, Bessac, Julie, Qian, Elizabeth

Inhaltsangabe: “… access to the underlying code or full model operators, i.e., non-intrusively. Traditional OpInf operates …”
Volltext