Unverified50% confidenceFactExact time
SFT is not installing a chat module into the model, but performing a large-scale reweighting of the probability distribution to increase the probability of the reference answer sequence.
1
Sources
50%
Confidence
Long-term
Relevance
10/31/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedSupervised Fine-Tuning (SFT) trains the model using 'prompt-response' example pairs, teaching the model to respond in a conversational format, thereby making outputs more coherent.74% similarUnverifiedSFT is an alignment method that updates model parameters using paired inputs and outputs and a cross-entropy loss, without relying on a reward model or preference comparisons.68% similarUnverifiedHybrid Resampling only requires transmitting the probability distribution of the shared vocabulary region between the device and server, greatly reducing the volume of communication data, and is mathematically proven to fully preserve the server LLM's original probability distribution for lossless inference.65% similarVerifiedDecision models do not generate text; instead, given an input state and a set of named questions, they return typed answers with probabilities in a single forward pass.65% similarUnverifiedLarge model fine-tuning engineers (SFT direction) are one of the fastest-growing areas of enterprise demand, because enterprises often perform SFT supervised fine-tuning and LoRA fine-tuning on existing models using internal data rather than training models from scratch.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1121443API
curl https://kongchang.com/api/v1/knowledge/claims/1121443MCP
get_claim(id=1121443)