CLASP model detects malicious tokens using XGBoost classifier, achieving 95.9% token-level F1 score.

Alexandre Le Mercier, Thomas Demeester, Chris Develder

IndexCache accelerates sparse attention by reusing cross-layer indices, reducing 75% of computations, achieving 1.82x speedup.

Yushi Bai, Qian Dong, Ting Jiang et al.

Introduced a Polish long-context encoder model handling up to 8192 tokens, significantly improving long-document task performance.

Sławomir Dadas, Rafał Poświata, Marek Kozłowski et al.

ComFree-Sim is a GPU-parallelized contact physics engine achieving near-linear scaling in contact-rich scenarios, with 2-3x throughput improvement.

Chetan Borse, Zhixian Xie, Wei-Cheng Huang et al.

Quantifies forgetting in generative models post-training using forward and reverse KL objectives, avoiding quality degradation.

Krishnakumar Balasubramanian, Shiva Prasad Kasiviswanathan

LifeSim simulates user cognition via BDI model to enhance personalized assistant evaluation.

Feiyu Duan, Xuanjing Huang, Zhongyu Wei

O3N framework achieves state-of-the-art performance on QuadOcc and Human360Occ benchmarks using polar-spiral topology for 360° spatial representation.

Mengfei Duan, Hao Shi, Fei Teng et al.

Proposed dynamic modeling and gravity compensation for dVRK-Si PSM, reducing joint errors by 68-84%.

Haoying Zhou, Hao Yang, Brendan Burkhart et al.

Introduced UniCAC benchmark to evaluate 24 algorithms under various optical aberrations.

Xiaolong Qian, Qi Jiang, Yao Gao et al.

Proposed a decentralized cooperative localization framework with asynchronous sensor fusion, achieving 34% RMSE reduction.

Nivand Khosravi, Niusha Khosravi, Mohammad Bozorg et al.

SNAP-V: A RISC-V SoC optimized for small-scale SNN inference, average synaptic energy 1.05 pJ.

Kanishka Gunawardana, Sanka Peeris, Kavishka Rambukwella et al.

EnTransformer combines Transformer with engression for superior multivariate probabilistic forecasting.

Rajdeep Pathak, Rahul Goswami, Madhurima Panja et al.

Modeling trial-and-error navigation using a sequential decision model of information scent under memory constraints.

Xiaofu Jin, Yunpeng Bai, Antti Oulasvirta

Stable Spike achieves dual consistency optimization via bitwise AND operations, enhancing SNN recognition performance under ultra-low latency by up to 8.33%.

Yongqi Ding, Kunshan Yang, Linze Li et al.

FedShare framework enhances recommendation performance through personalized data sharing and unlearning.

Liang Qu, Jianxin Li, Wei Yuan et al.

Quantum mechanical framework for quantization-based optimization reveals quantum tunneling aids in escaping local minima; experiments show superior performance over traditional algorithms.

Jinwuk Seok, Changsik Cho

OneRec-V2 achieves 49% latency reduction and 92% throughput increase via FP8 quantized inference.

Yi Su, Xinchen Luo, Hongtao Cheng et al.

MDER-DR framework enhances multi-hop QA with entity-centric summaries, achieving 66% improvement.

Riccardo Campi, Nicolò Oreste Pinciroli Vago, Mathyas Giudici et al.

COMIC system uses LLM critics to generate sketch comedy videos near professional quality.

Susung Hong, Brian Curless, Ira Kemelmacher-Shlizerman et al.

NeFTY achieves high-accuracy 3D thermal diffusion reconstruction using a differentiable physics framework, significantly improving defect localization.

Tao Zhong, Yixun Hu, Dongzhe Zheng et al.