Unverified50% confidenceFactExact time
The author used Tinker's LoRA training API to fine-tune Qwen3.5-4B on 5,755 Open Food Facts entries; the training set contains 5,000 ordinary products, 550 refusal samples, and 205 control samples.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/18/2026
First Seen
Valid until: 1/16/2027
Sources
Related Entities
Related Claims
Unverified后训练量化通过少量代表性校准数据（通常128-1024条样本）前向传播收集各层激活统计特征，常用校准数据集包括WikiText-2、C4和Pile子集65% similarUnverifiedMiniCPM5-2B open-sourced its training data, including approximately 500,000 agent samples and over 80,000 reinforcement learning samples.64% similarUnverified智谱GLM-5.3基于743B（7430亿参数）基座进行后训练，主打编程与智能体两大核心场景64% similarUnverifiedLoRA训练中rank值通常建议从4-16或4-32的安全区间开始实验，过高的rank值会导致模型记忆训练细节而非抽象人物特征62% similarUnverifiedTev proposed Tev-Gym, reportedly the first long-horizon agent training dataset with verifiable, dense, progressive rewards, covering 7 domains and a total of 1170 validated task environments.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1060641API
curl https://kongchang.com/api/v1/knowledge/claims/1060641MCP
get_claim(id=1060641)