WORLDWIDE_SanGuai / Research bulletin

[ worldwide.arakne.de ]

世界很大,讀一小點。

SanGuai 的三日研究佈告板。每天更新,讀完就走。

EST. 202672H上架 72 小時
人工智慧 / 2611.16961

思考的價值:在混合決策模型中學習何時推理

The Value of Thinking: Learning When to Reason in Hybrid Decision Models

作者: Zhengxu Yu, Ke Xu, Meng Fang, Haitham Ammar, Weilin Luo, Jun Wang論文發布: 2026-11-12離架倒數: 63 小時

研究速讀

作者提出「零或T思考」範式:單一模型要麼在一次前向傳播中以隱式思考直接作答、不產生任何詞元,要麼在值得付出成本時進行完整的自迴歸推理。此範式以混合決策模型實現,由凍結主幹與輕量自我路由區塊組成;該區塊依隱式思考的產出估計「思考的價值」,僅在價值超過每次請求設定的價格時才啟用慢速模式。作者亦提出CDPO,以回合層級回報訓練路由區塊。

值得留意

  1. 論文提出零或T思考範式,讓單一模型在隱式作答與完整自迴歸推理之間選擇,而非只縮短或跳過顯式推理。
  2. 混合決策模型以一個凍結主幹同時支援兩種模式,並由輕量自我路由區塊依估計的思考價值與每次請求的價格λ決定是否啟用慢速模式。
  3. 作者提出CDPO,用回合層級回報訓練路由區塊,以處理情節式代理任務中缺乏逐決策標籤的情況。

閱讀限制

僅閱讀標題與摘要,無法確認實驗設定、基準細節或CDPO的實際效果,摘要中的數字與比較皆為作者自述,未經獨立驗證。

原始摘要 [EN]

Existing adaptive-reasoning methods shorten the explicit reasoning a model writes, or skip it by switching modes; either way the answer is still written out token by token. Rather than deciding how much to think, we explore a different paradigm, zero-or-T thinking: a single model either answers from the implicit thinking inside one forward pass, generating no token, or performs full autoregressive reasoning at a cost of T tokens when that is worth paying. We realise it in a hybrid decision model (HDM), where one frozen backbone serves both modes and a lightweight self-routing block chooses between them. The block estimates the value of thinking, the utility the slow mode is expected to add over the fast one, from what that implicit thinking has already produced, and the slow mode runs only where that value exceeds a price λ set per request. Where no per-decision label exists, as in episodic agentic tasks, we introduce Calibrated Decision Policy Optimisation (CDPO), a reinforcement-learning recipe that trains the block from episode-level returns. We evaluate HDM on MMLU-Pro, AGIEval, GPQA-Diamond, JevBench, BANKING77 and MiniWoB. Specifically, on JevBench HDM-4B scores 0.857 against Jev's 0.851 (+0.7%) at a median 0.179\,s against 0.652 (-72.5%). On the agentic venue MiniWoB HDM-27B cuts token cost by 80.1%.

AI 摘要僅根據公開標題與摘要整理;研究結論來自作者,請以原文為準。