Unverified50% confidenceFactExact time
When the model is uncertain about a word, CuriousLM displays an underline beneath that word; clicking it shows candidate words and allows switching the generation path.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/2/2026
First Seen
Valid until: 1/31/2027
Sources
Related Entities
Related Claims
UnverifiedIn traditional evaluations, models can rely on parametric knowledge or semantic cues in the compressed text to guess the answer even when the key evidence has been removed, a phenomenon called "shortcut learning."67% similarUnverifiedThe study constructs a theory-grounded taxonomy that distinguishes curiosity-driven questions from general model questions: the former point toward unknown domains, while the latter are closer to customer-service-style confirmation or task clarification.66% similarVerifiedWhen a prompt is entered, the model predicts the most likely next token based on context, rather than truly understanding the semantics.65% similarUnverifiedWhen a large model finds insufficient information in the prompt, the Ask User Question mechanism allows it to proactively ask the user questions to fill contextual gaps.64% similarUnverified知识图谱以工具（tools）形式暴露给前沿模型，而非直接塞进 prompt，使模型可按需多次调用查询64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1126713API
curl https://kongchang.com/api/v1/knowledge/claims/1126713MCP
get_claim(id=1126713)