Unverified50% confidenceFactExact time
EmoHarness is an external framework that guides a frozen small model to use multimodal evidence by organizing prompts, settings, and verification code, without updating model weights.
1
Sources
50%
Confidence
Long-term
Relevance
11/10/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedEmoGenie is accompanied by an evaluation model, EmoProbe.71% similarUnverifiedEmbedding Gemma 2 is an open-source multimodal retrieval model that supports fine-tuning with one's own data.70% similarUnverifiedEmbedding模型微调通常采用对比学习范式,典型方法是构造查询-正例文档-负例文档三元组,通过InfoNCE或Triplet Loss等损失函数优化69% similarUnverifiedOcclusion is a class of perturbation-based post-hoc explanation methods that measures importance by systematically occluding different parts of the input and observing changes in the model's output probabilities.67% similarUnverifiedEmbedding model fine-tuning typically adopts a contrastive learning framework, training the model to bring positive sample pairs closer together and push negative sample pairs apart by constructing positive and negative sample pairs.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1155385API
curl https://kongchang.com/api/v1/knowledge/claims/1155385MCP
get_claim(id=1155385)