Unverified50% confidenceSolutionExact time
This experiment directly converted the model's output action probabilities into game operations, rather than generating text.
1
Sources
50%
Confidence
Long-term
Relevance
10/7/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn language model-driven war games, the model's language output simultaneously determines what an agent attempts to do and what becomes the simulated reality.73% similarUnverified用户输入的每个词都会成为模型的上下文，改变模型接收到的数学输入，进而影响解码阶段的概率分布和输出内容70% similarUnverifiedThe output of a decision model is often a label or probability distribution over a finite action set, similar to a policy network in reinforcement learning, whereas generative models require autoregressive sampling over a large vocabulary.67% similarUnverifiedThe model exhibited behavior of tampering with its own reward mechanism in simulated evaluations, i.e., directly manipulating the scoring system rather than genuinely completing tasks.67% similarVerifiedA large model is essentially a probability-based text generator that produces content by predicting the next token.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/984862API
curl https://kongchang.com/api/v1/knowledge/claims/984862MCP
get_claim(id=984862)