Unverified50% confidenceOpinionExact time
Existing LLM-based ToM methods usually treat ToM as a subtask of language reasoning, rely on explicit cues in text, and lack independent modeling of the internal structure and temporal evolution of mental states.
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedLarge language models exhibit a "knowing but not acting" phenomenon in theory of mind (ToM) reasoning tasks: they can correctly infer a character's false belief, but when predicting that character's subsequent action, they give an answer based on objective facts rather than the false belief.76% similarUnverifiedTemporal multimodal large language models (TS-MLLMs) leverage the reasoning capabilities of large language models to handle question-answering tasks, but struggle to genuinely capture dynamic temporal patterns, typically providing implicit reasoning that lacks explicit explanation of the logic behind conclusions.68% similarUnverifiedLLM推理框架正在从单纯的文本生成引擎演进为智能体运行时67% similarUnverified思维链提示有效的原理是：LLM是逐词生成的自回归模型，先生成推理过程可作为后续推断的工作记忆，降低直接跳结论的错误率67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1065894API
curl https://kongchang.com/api/v1/knowledge/claims/1065894MCP
get_claim(id=1065894)