Zapis dokaza metode

Direct Preference Optimization

Direct Preference Optimization (DPO) is a training method introduced by Rafailov et al. in 2023 that aligns language models with human preferences without requiring an explicit reward model. By directly optimizing for preference pairs (better response vs worse response), DPO simplifies the training pipeline compared to reinforcement learning from human feedback (RLHF).

Sources recorded, not reviewed

Izvorni zapis

Citati kopirani doslovno iz izvornog zapisa metode. Ne impliciraju nikakvu provjeru na razini tvrdnje.

Direct Preference Optimization: Your Language Model is Secretly a Reward Model

Taksonomski zapis metode · ml-model / deep-learning

Rafailov, R., Sharma, A., Mitchell, E., Manning, C. D., Ermon, S., & Finn, C. (2023). Direct preference optimization: Your language model is secretly a reward model. arXiv preprint arXiv:2305.18290. · URL

Otvori cijelu metodu

Uređene tvrdnje

Tvrdnje pohranjene u knjigu dokaza, svaka s vlastitom procjenom.

Nema uređenih tvrdnji

Ovaj prikaz ne izmišlja procjenu tvrdnje kada knjiga dokaza nema nijednu.

Povezane metode

Generirano iz grafa metode i prikazano kao strojno predložene relacije — ne implicira se nikakva tvrdnja dokaza.

Same method familyLatent Diffusion Modelsmachine-suggested · Relational suggestion, not evidence.Same method familyMamba (State Space Model)machine-suggested · Relational suggestion, not evidence.Same method familyMasked Autoencodersmachine-suggested · Relational suggestion, not evidence.Same method familyQLoRAmachine-suggested · Relational suggestion, not evidence.