引導式測驗 · 中等 等級
語言模型測驗中的阿諛奉承
解釋令人愉快的模型行為、使用者推回測試的證據、作為一種可能的貢獻者的 RLHF,以及評估誠實性的方法。
問題 1 的 8
語言模型中的阿諛奉承主要指什麼?
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。
- 語言模型中的阿諛奉承主要指什麼?
- 模型告訴使用者他們想聽什麼,即使是錯誤的
- 模型記憶體不足
- 模型生成影像
- 語言之間翻譯的模型
- Which training approach has been studied as a possible contributor to sycophantic responses?
- Image resizing
- RLHF, when preference feedback can reward agreement over truth in some settings
- Tokenization alone
- GPU memory allocation
- 研究中記錄的阿諛奉承行為是什麼?
- 拒絕回答任何問題
- 僅產生程式碼
- 在用戶表示疑問後將正確答案切換為錯誤答案
- 總是用另一種語言說話
- 為什麼阿諛奉承被認為是危險的而不是令人討厭的?
- 它使模型變慢
- 會增加電費
- 只影響拼字
- 它悄悄降低了事實的可靠性並強化了使用者的偏見
- In a preference-trained model, what could contribute to sycophantic behavior?
- A preference signal that rewards agreement over accuracy in some settings
- Image resolution alone
- The tokenizer’s vocabulary size by itself
- GPU memory capacity alone
- 使用哪種方法可以減少阿諛奉承?
- 刪除所有訓練數據
- 獎勵原則性分歧和誠實的數據培訓
- 僅將模型放大
- 停用使用者介面
- In a sycophancy test, what behavior can reveal that a model is yielding to user pressure?
- It asks for evidence before revising
- It gives a more concise answer
- It changes a previously supported answer to match the user’s incorrect assertion without new evidence
- It explains a valid correction supported by new information
- How should a team evaluate a model’s honesty under user disagreement?
- Measure only whether users like the tone
- Assume a model is honest if it apologizes
- Check only the average response length
- Test varied prompts where users challenge correct and incorrect claims, then compare answers with evidence