SelanjutnyaPanduan berikutnya
Pengambilan Sampel Khas
Bahasa AI
PANDUAN AI Bahasa
Top-k and nucleus (top-p) sampling are stochastic decoding methods that truncate a model’s next-token distribution before drawing a token.
Dengarkan panduan iniDiceritakan oleh AI · ~2 min
Temperature rescales logits and changes distribution sharpness; none of these settings guarantees factual accuracy.
Holtzman et al. identify a gap between likelihood training and open-ended generation: likelihood-maximizing decoding can produce bland and repetitive text in their experiments. Nucleus sampling retains a dynamic set of tokens containing most probability mass and samples from it, trimming the unreliable tail. Top-k instead fixes the number of candidates. Later analysis explains properties and limitations of truncation heuristics; their findings are scoped to the studied models and tasks. For a no-sampling baseline, greedy decoding selects the highest-scoring next token directly. This removes token-sampling randomness for a fixed model and input; it does not guarantee correct code, and deployment-level reproducibility can also depend on system details.
Top-k keeps a fixed number of high-probability tokens at a step.
Top-p keeps the smallest probability-ranked set whose cumulative mass reaches the chosen threshold.
Temperature changes the relative sharpness of the distribution before sampling.
When evaluating open-ended outputs, compare decoding settings on diversity, coherence and repetition rather than assuming one setting is best.
Choose decoding settings by task-specific evaluation and record the model, prompt and sampling parameters for reproducibility. Assess diversity and coherence alongside task success. The cited papers do not establish a universal default or guarantee that sampling prevents repetition or errors.
Buletin gratis
Tiga cerita AI terverifikasi setiap pagi hari kerja, ditulis dalam bahasa Inggris sederhana. Gratis selamanya, tanpa iklan.
Satu email setiap hari kerja. Berhenti berlangganan dalam satu klik. Kami tidak pernah menjual atau membagikan alamat Anda.
Uji diri Anda sendiri
Umpan balik instan untuk setiap jawaban, dan sertifikat yang dapat dibagikan dengan ID yang dapat diverifikasi setelah Anda lulus kursus.
Mendukung pendidikan AI gratis. AI Understanding adalah organisasi nirlaba 501(c)(3) — tanpa iklan, tanpa paywall, selamanya. Berikan sumbangan
Top-k and nucleus (top-p) sampling are stochastic decoding methods that truncate a model’s next-token distribution before drawing a token. Temperature rescales logits and changes distribution sharpness; none of these settings guarantees factual accuracy.
Top-k memotong distribusi ke sejumlah k dari token yang paling mungkin, melakukan normalisasi ulang, dan mengambil sampel dari token tersebut.
Top-p menyimpan grup terkecil dari token teratas yang probabilitas kumulatifnya mencapai ambang batas p, lalu mengambil sampel di antara token tersebut.
Kumpulan kandidat Top-p menyusut ketika model yakin dan tumbuh ketika tidak pasti, tidak seperti jumlah tetap top-k.
Suhu yang lebih rendah memusatkan probabilitas pada token teratas (lebih deterministik); suhu yang lebih tinggi meratakannya (lebih beragam).
Holtzman et al. report that sampling from the dynamic nucleus produced more diverse generations while retaining fluency and coherence in their evaluated setups; this is not a universal guarantee.
Teruslah belajar
Panduan lainnya dipilih untuk topik ini
SelanjutnyaPanduan berikutnya
Pengambilan Sampel Khas
Bahasa AI