master
- Add 15-token length floor to SFT samples, with explicit reference to Appendix A.3 / Figure 5 (ifd_length_grid) - Short replies (<10 tokens) show both high per-token perplexity (L_uncond ~6-8, PPL ~400-3000 vs long replies ~2-3, PPL ~7-20) and wide variance in L_cond (span 0-17.5), which distorts downstream IFD-based difficulty estimates.
Description
No description provided
6.3 MiB
Languages
TeX
100%