Unverified50% confidenceFactExact time

This cross-encoder has 307 million parameters and was trained on thirteen Russian tasks.

1

Sources

50%

Confidence

Long-term

Relevance

11/6/2026

First Seen

Sources

Label-Wording Augmentation: Effective for Unfamiliar Words, Ineffective for Abbreviations and Code

Related Entities

Related Claims

Unverified现代大语言模型的预训练数据集通常包含数万亿 token，涵盖 GitHub 代码仓库、Stack Overflow、Linux 文档等61% similar UnverifiedCodeBERT的训练数据来自GitHub上超过600万个函数的代码-注释配对61% similar VerifiedCodex's training data covers billions of lines of publicly available code on GitHub.60% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/1142292

API

curl https://kongchang.com/api/v1/knowledge/claims/1142292

MCP

get_claim(id=1142292)