TAC-CoT:透過交錯明確 token 實現工具賦能的隱式思維鏈
TAC-CoT: Tool-Empowered Implicit Chain-of-Thought via Interleaving Explicit Tokens
TAC-CoT 是一種工具賦能的自適應隱式 CoT 框架,以精簡的潛在推理片段取代明確推理軌跡,並只在工具呼叫、觀察及最終答案中保留文字 tokens。在智能體數學及問答基準測試中,準確率最高提升 8.8 points,模型 forward passes 最多減少 6.8x。分析亦顯示,TAC-CoT 在單工具及多工具設定中均改善工具使用質素。
Published on Nov 1
Authors:
,
,
Abstract
TAC-CoT replaces explicit reasoning traces with compact latent segments while preserving textual tokens for tool interactions, improving agentic accuracy and reducing computation.
LLM-based agentic systems are increasingly combined with CoT reasoning to solve tasks that require accurate reasoning, external knowledge, and multi-step deliberation. Yet, explicit CoT incurs substantial generation overhead, while implicit CoT, which replaces textual rationales with compact continuous tokens, does not naturally support the explicit communication interface required for agentic tool-use. We propose TAC-CoT, a tool-empowered adaptive implicit CoT framework that replaces explicit reasoning traces with compact latent reasoning segments while retaining textual tokens only for tool calls, observations, and final answers. This interleaved design enables agentic tool-use without verbalizing every intermediate reasoning step. Experiments on agentic math and QA benchmarks show that TAC-CoT improves accuracy by up to 8.8 points while reducing model forward passes by up to 6.8x. Furthermore, analyses show that TAC-CoT improves tool-use quality in both single- and multi-tool settings.
View arXiv page View PDF Project page Add to collection
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2611.01926 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2611.01926 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2611.01926 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co