Unverified50% confidenceFactExact time
The core mechanism of MPT is to let agents express their willingness to participate through bidding; bids determine who takes action and how the collective outcome provides local training credit, thereby enabling joint learning of action strategies and bidding behavior without a central coordinator.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
Unverified"Market-based Post-training" (MPT) is a framework that lets a decentralized swarm of agents simultaneously learn action strategies and self-orchestration.77% similarUnverifiedMPT's comparative experiments show that under the same data and compute budget, letting agents learn strategies separately and coordinate through bidding not only achieves better task performance but also remains stable when the number of agents changes.71% similarUnverifiedThe research team evaluated MPT on three tasks—embodied crafting, robot navigation, and deep research—and the results show that its performance surpasses monolithic post-training, and that agents' participation patterns evolve with learning.69% similarUnverifiedBy constructing supervisory signals from the data itself, self-supervised learning has become the core training strategy for large-scale pre-trained models such as BERT, GPT, and MAE.66% similarUnverifiedMPT's evaluation is currently still limited to specific tasks, and the generalization ability of the bidding mechanism in more open environments, communication costs, and comparisons with other coordination methods still await further research.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1041934API
curl https://kongchang.com/api/v1/knowledge/claims/1041934MCP
get_claim(id=1041934)