Unverified50% confidenceFactExact time
SeTPO stands for Selective Teacher-guided Policy Optimization, a method for tool learning in small language models.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDecision models are typically based on reinforcement learning or preference-based optimization methods, with the direct optimization target of selecting the optimal action in the environment, whereas LLMs have language prediction as their core task and decision-making capability is an emergent byproduct.67% similarUnverifiedForward KL tends to make the student policy cover all modes of the teacher policy, learning comprehensively but possibly not precisely enough; reverse KL tends to make the student policy concentrate on the high-probability regions of the teacher policy, learning precisely but possibly missing certain modes.66% similarUnverifiedOn-policy distillation (OPD) has the student model generate its own trajectories, and then the teacher model provides supervision signals on those trajectories.64% similarUnverifiedProximal Policy Optimization (PPO) is the primary policy gradient algorithm tool for robot reinforcement learning.64% similarUnverifiedTeacher forcing introduces the exposure bias problem: during training the real target token is fed as input, while during inference the model's own predicted token is used, and prediction errors accumulate and amplify in subsequent steps.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1023679API
curl https://kongchang.com/api/v1/knowledge/claims/1023679MCP
get_claim(id=1023679)