引導式測驗 · 中等 等級
數學推理測驗過程監督
使用 PRM800K 標籤、候選人選擇、Math-Shepherd 監督和步驟分數聚合檢查流程和結果回饋。
問題 1 的 8
與結果監督相比,過程監督在何處提供回饋?
預覽每個問題8 問題
當您參加測驗時,就會出現答案和解釋。
- 與結果監督相比,過程監督在何處提供回饋?
- Only at the physical GPU temperature
- At intermediate reasoning steps rather than only the final result
- Only at the document filename
- Only at an increased vocabulary size
- 為什麼只注重結果的監督會對數學問題產生誤導?
- 計算速度太慢
- 它無法處理小數
- 每個問題都需要人工評分
- 模型可以透過有缺陷的步驟得出正確的答案,而這些步驟碰巧會被抵消
- OpenAI 與「讓我們逐步驗證」一起發布了什麼?
- PRM800K,約 800,000 個步驟級標籤的資料集
- 全新 GPU 架構
- GSM8K測試儀
- 國際象棋引擎
- 在使用過程獎勵模型的 Best-of-N 數學選擇中,驗證器做了什麼?
- Guarantees that a correct candidate was generated
- Changes every tokenizer token into a proof axiom
- Scores generated candidate solutions under the chosen aggregation rule
- Replaces the original problem with a translation
- 什麼方法可以減少對昂貴的人工步驟標籤的需求?
- 完全禁用獎勵模型
- 自動估計每個步驟值的蒙特卡羅卷展欄(例如,Math-Shepherd)
- 僅使用最終答案
- 更多語言培訓
- What can low step-level verifier scores usefully indicate to someone reviewing a solution?
- Steps that merit closer inspection
- An infallible location of the first actual error
- A mathematical proof that all other steps are correct
- The exact temperature used to sample every token
- For step scores 0.9 and 0.8, what is the solution score under a product aggregation rule?
- 0.8
- 0.85
- 1.7
- 0.72
- Which data-selection strategy did Let’s Verify Step by Step study to improve labeling efficiency?
- Choosing examples only by alphabetical filename order
- Removing every incorrect-answer candidate from the labeling pool
- Active learning that selected informative candidate solutions
- Using an unlabeled test answer as the final human label