Unverified50% confidenceFactExact time
Even the most advanced current models have a notable gap in factual detail accuracy on summarization tasks.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/11/2026
First Seen
Valid until: 12/10/2026
Sources
Related Entities
Related Claims
Unverified模型压缩技术（量化、剪枝、知识蒸馏）会不成比例地损害模型的事实准确性，量化后模型在事实性问答任务上的幻觉率可能显著上升75% similarUnverifiedLarge models output the most likely next word rather than precise retrieval of facts, which leads to hallucinations in domains with sparse coverage in the training data.73% similarUnverifiedCurrent mainstream large models excel at handling isolated, well-bounded small tasks, but expose shortcomings of incomplete understanding and loss of detail when faced with highly coupled, large-context legacy systems.73% similarUnverifiedPure large language models have two major deficiencies: a knowledge cutoff problem and a hallucination problem.72% similarUnverifiedTraditional frame-sampling methods suffer from significant information loss, lack of temporal reasoning, and high computational costs.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/904924API
curl https://kongchang.com/api/v1/knowledge/claims/904924MCP
get_claim(id=904924)