Unverified50% confidenceEventExact time
The study conducted experiments on three tasks: screening harmful requests to large language models, qualitative text annotation, and data annotation.
1
Sources
50%
Confidence
Long-term
Relevance
10/23/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe study tested the introduction of an auxiliary language identification task as an intervention to strengthen language differentiation capability.74% similarUnverifiedThe study provides behavioral-level evidence that "bilingual models capture some aspects of cross-linguistic influence."71% similarUnverifiedThe training corpora of large language models may include unpublished academic preprints and research notes, raising concerns about data contamination.70% similarUnverifiedVocalGrad serves as a reminder: high scores of multimodal models on semantic tasks may mask gaps in perceptual ability, and when models rely on textual priors or linguistic reasoning to guess answers, benchmarks will overestimate their true ability.69% similarUnverifiedWith the widespread use of large language models in academic writing, citation-related academic misconduct issues have become increasingly prominent.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1088411API
curl https://kongchang.com/api/v1/knowledge/claims/1088411MCP
get_claim(id=1088411)