Unverified50% confidenceBenchmarkExact time
On BrowseComp-Plus, attribution-guided SFT matches the effect of full-turn exact-context SFT using only about 30% of the training tokens.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/20/2026
First Seen
Valid until: 1/18/2027
Sources
Related Entities
Related Claims
UnverifiedThe planner was trained via supervised fine-tuning (SFT) on outcome-selection trajectories from 140 training questions and 28 development questions.77% similarUnverifiedFor classification tasks with discrete and closed label spaces, the training signal from SFT is already clear enough, and introducing reinforcement learning would instead increase the difficulty of hyperparameter tuning and training instability74% similarUnverifiedPre-training and SFT provide only positive supervision, whereas DPO and adjacent RL techniques provide the leverage of negative supervision, which can reduce the probability of erroneous sequences.70% similarVerifiedLarge model training goes through three core stages: pre-training, supervised fine-tuning (SFT), and preference alignment.70% similarUnverifiedIn the post-training phase, Motif adopts a "failure-driven SFT" strategy: identifying model weaknesses through benchmark testing and internal scenarios, then generating targeted datasets for training.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073979API
curl https://kongchang.com/api/v1/knowledge/claims/1073979MCP
get_claim(id=1073979)