Web This site Trusted
19 result(s) for “okapi bm25 site hotchips pp site github com site of de site huggingface co site does site:doi.org” · 1026 ms
Scope: Nishi’s own index (our crawl + Common Crawl) — not the whole live web. Some sites block all automated crawlers (anti-bot walls), so their pages may be absent here even though we never filter lawful content.
Freshness: newest result fetched 26 d ago · median age of these results 26 d · 1 of 19 carry a fetch time
In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript
from the top result below
https://dx.doi.org/10.1038/s41586-026-11044-y
In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript
https://doi.org/10.5281/zenodo.21722356
com/Sisyphbaous-DT-Project/open-qingyi - Weights: https://huggingface
https://doi.org/10.48550/arXiv.2509.22186
Jiaqi Wang, Bo Zhang, Lei Bai, Pei Chu, Weijia Li, Jiang Wu, Lijun Wu, Zhenxiang Li, Guangyu Wang, Zhongying Tu, Chao Xu, Kai Chen, Yu Qiao, Bowen Zhou, Dahua Lin, Wentao Zhang, Conghui He
View a PDF of the paper titled MinerU2
https://doi.org/10.48550/arXiv.2509.02523
02523] Flavors of Moonshine: Tiny Specialized ASR Models for Edge Devices
Skip to main content
arXiv is now an independent nonprofit
https://doi.org/10.48550/arXiv.2311.12983
We posit that the advent of Artificial General Intelligence (AGI) hinges on a system's capability to exhibit similar robustness as the average human does on such questions
https://doi.org/10.48550/arXiv.2006.01311
Simulation tests and mathematical proofs show the efficiency of the algorithm
https://doi.org/10.48550/arXiv.2508.19205
Thus, VibeVoice can synthesize long-form speech for up to 90 minutes (in a 64K context window length) with a maximum of 4 speakers, capturing the authentic conversational ``vibe'' and surpassing open-source and proprietary dialogue models
https://doi.org/10.48550/arXiv.2006.01758
It has realised a great number of application scenarios, and improved workflow of many domains
https://doi.org/10.48550/arXiv.2006.01402
We propose a priority-based background scheduler that learns this repetitive pattern and allows storage systems to maintain peak performance and in turn meet service level objectives (SLOs) while supporting a number of data services
https://doi.org/10.48550/arXiv.2212.07841
for further improving the quality of dense vectors
https://doi.org/10.48550/arXiv.2508.02739
02739] Kronos: A Foundation Model for the Language of Financial Markets
Skip to main content
arXiv is now an independent nonprofit
https://doi.org/10.48550/arXiv.2211.01910
However, task performance depends significantly on the quality of the prompt used to steer the model, and most effective prompts have been handcrafted by humans
https://doi.org/10.48550/arXiv.2006.01957
Rodriguez, Rajkumar Buyya
View a PDF of the paper titled Workflow-as-a-Service Cloud Platform and Deployment of Bioinformatics Workflow Applications, by Muhammad H
https://doi.org/10.48550/arXiv.2211.10435
Much of this success can be attributed to prompting methods such as "chain-of-thought'', which employ LLMs for both understanding the problem description by decomposing it into steps, as well as solving each step of the problem
https://doi.org/10.48550/arXiv.2608.03063
Decisions at WeChat Pay, by Guilin Li and 4 other authors
View PDF HTML (experimental)
Abstract:Merchant risk control at large payment platforms screens tens of millions of merchants daily, where false positives harm legitimate merchants and false negatives leave harmful activity undetected
https://doi.org/10.48550/arXiv.2508.16279
Yaliang Li, Bolin Ding, Jingren Zhou
View a PDF of the paper titled AgentScope 1
https://doi.org/10.1145%2F3397271.3401075
While remarkably effective, the ranking models based on these LMs increase computational cost by orders of magnitude over prior approaches, particularly as they must feed each query-document pair through a massive neural network to compute a single relevance score
https://doi.org/10.48550/arXiv.2309.17428
In addition, our in-depth analysis reveals that: (1) consistent performance improvement can be achieved by scaling up the number of tools and the capability of the backbone models; (2) each component of our approach contributes to the performance gains; (3) the created tools are well-structured and reliable with low
https://doi.org/10.48550/arXiv.2307.16789
This is in contrast to the excellent tool-use capabilities of state-of-the-art (SOTA) closed-source LLMs, e