Consensus Mechanism
When synthesizing answers, LLMs tend to adopt information that multiple independent sources agree on; if sources conflict, the model may hedge. It is one of the core mechanisms of answer synthesis in AI search.
Timeline (last 90 days)
The reward-free filtering uses a consensus mechanism (sampling takeovers multiple times and keeping segments with consistent outcomes) or LLM judging (a large model judges whether the takeover is reasonable) to filter out low-quality trajectories, and the entire process does not rely on task success signals.
The consensus mechanism refers to the tendency of an LLM, when generating an answer, to adopt information consistently recognized by multiple sources; if a brand is described in a similar way on Reddit, industry forums, news websites and official documents, the model is more likely to adopt that description.
All Facts (2)
The reward-free filtering uses a consensus mechanism (sampling takeovers multiple times and keeping segments with consistent outcomes) or LLM judging (a large model judges whether the takeover is reasonable) to filter out low-quality trajectories, and the entire process does not rely on task success signals.
50%UnverifiedThe consensus mechanism refers to the tendency of an LLM, when generating an answer, to adopt information consistently recognized by multiple sources; if a brand is described in a similar way on Reddit, industry forums, news websites and official documents, the model is more likely to adopt that description.
50%