Web This site Trusted
30 result(s) for “okapi bm25 site hotchips pp site github com site no not gov site go de site:arxiv.org” of 37 matched · 733 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: no result on this page carries a fetch time yet (indexed before the fetch epoch was stored beside the doc)
The query asks the model to recall the topic of an earlier conversation about Tesla; the gold answer is anchored in one specific session about Pooja’s dig site
from the top result below
https://arxiv.org/html/2608.02613v1
4 Permission-Aware Access: Why Non-Leaks Are Not Refusals
A
https://arxiv.org/html/2608.03048v1
com/JetAstra/PI-Mem
Introduction
The ability to reason over long contexts is increasingly important for large language models (LLMs), supporting applications such as long-document question answering (Gemini Team 2025; Yang et al
https://arxiv.org/html/2608.02668v1
On nanoGPT, all three methods outperform existing lightweight deep connection schemes, and the best validation loss is attained at finite pp, indicating that the exponential map is not the preferred retraction for spherical residual streams but merely one end of a spectrum
https://arxiv.org/html/2407.16741v3
com/All-Hands-AI/OpenHands
Slack http://bit
https://arxiv.org/html/2608.03063v1
8 pp in a production fraud detector serving billion-scale transaction traffic
https://arxiv.org/html/2608.02918v1
Biscainho are with SMT, DEL/Poli & PEE/COPPE, Federal University of Rio de Janeiro, Brazil (corresponding e-mail: wallace
https://arxiv.org/list/cs.AI/new
Skip to main content
System maintenance August 4th and 5th Learn more ×
Search Submit Donate Log in
Search arXiv
Press Enter to search · Advanced search
Artificial Intelligence
New submissions
Cross-lists
Replacements
See recent
https://arxiv.org/html/2501.13956v1
com
Abstract
We introduce Zep, a novel memory layer service for AI agents that outperforms the current state-of-the-art system, MemGPT, in the Deep Memory Retrieval (DMR) benchmark
https://arxiv.org/html/2608.03011v1
In many real-world application scenarios, however, the acoustic environment present in the speech signal is not merely an undesired distortion, but rather constitutes a desired auditory attribute of the synthesized speech
https://arxiv.org/html/2607.22511v2
Because a machine-checked proof establishes only that a formal statement follows from its assumptions—not that the statement faithfully captures the intended scientific claim—the pipeline augments kernel verification with a statement audit that compares each formal theorem against the informal claim it is intended
https://arxiv.org/html/2608.02616v1
2 Related Work
De-identification and text anonymization
https://arxiv.org/list/cs.CL/new
Prior work enables one-way, scholar-oriented translation from Akkadian to English, but offers no path in the reverse direction: non-specialist users cannot compose new content in cuneiform, and therefore remain passive consumers of ancient culture rather than active participants
https://arxiv.org/html/2608.02609v1
It is not a barrier between two living communities that might learn each other’s languages, but a one-way temporal asymmetry: ancient texts can be partially decoded for modern readers, but modern users have no productive channel into the ancient script
https://arxiv.org/html/2509.22186v2
\metadata
[* Equal contribution ✉ Corresponding author ‡\ddagger Project leader] \correspondenceConghui He, \metadata[Code:]https://github
https://arxiv.org/html/2210.00131v4
Code and open-source demos available at https://github
https://arxiv.org/html/2402.03300v3
com/deepseek-ai/DeepSeek-Math
Abstract
Mathematical reasoning poses a significant challenge for language models due to its complex and structured nature
https://arxiv.org/list/cs.IR/new
KGD improves over strong pretrain-transfer baselines by 4-12% on eight public benchmarks and sustains its advantage over a 90-day production stream where baselines show no gains
https://arxiv.org/html/2608.03372v1
Hedging and attribution are open classes: there is no complete list of ways to signal “I am not sure” or “someone told me,” so no list can be finished, and every list stalls at what its author thought of
https://arxiv.org/html/2607.26245v1
No strategy iteration survived fees and spread; out-of-sample, the strongest published model does not beat, and slightly underperforms the probability already implied by Polymarket’s own order book (Section 10
https://arxiv.org/html/2608.02612v1
Our code and dataset are available at https://github
https://arxiv.org/html/2608.00785v1
com
Abstract
Content discovery systems depend on relevance judgment for search quality evaluation, but human annotation faces inter-annotator disagreement and scaling costs
https://arxiv.org/html/2608.03446v1
To our knowledge, no prior work has inspected the impact of CLA scores on translation performance
https://arxiv.org/html/2407.17789v2
Last but not least, we provide a web-based interface for conveniently monitoring and managing a large number of agents that might deploy across multiple devices
https://arxiv.org/html/2403.08299v1
The landscape of software development has witnessed a paradigm shift with the advent of AI-powered assistants, exemplified by GitHub Copilot
https://arxiv.org/html/2508.16279v1
Applications
Report GitHub Issue
×
Title:
Content selection saved. Describe the issue below:
Description:
Submit without GitHub Submit in GitHub
arXiv is now an independent nonprofit! Learn more ×
Back to arXiv
Why HTML? Report
https://arxiv.org/html/2608.03077v1
This process-output alignment breaks the credit-assignment bottleneck in reasoning-augmented MT, allowing the model to optimize not only final translation quality which translation decisions support domain-faithful translation
https://arxiv.org/html/2508.19205v1
com/microsoft/VibeVoice
Hugging Face: microsoft/VibeVoice
Demo: aka
https://arxiv.org/html/2407.02518v2
com
Abstract
Large language models (LLMs) for code are typically trained to align with natural language instructions to closely follow their intentions and requirements
https://arxiv.org/abs/2608.03883
Because no annotated training data were released for this first edition of the task, we relied on zero-shot experimentation, manual validation, and unsupervised or weakly guided retrieval pipelines
https://arxiv.org/html/2608.03486v1
While linear SVAR models are useful in many applications, they are not able to capture nonlinear features of economic dynamics