Reading Comprehension on QuALITY derived transfer
42.5Performance (32k Context)Still
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| StillDecoding=deterministic greedy generation, max_new_tokens=192, repetition penalty=1.2, no-repeat 4-gram blocking=true2026.06 | 42.5 | 42 | 33.8 | |
| KV-DistillDecoding=deterministic greedy generation, max_new_tokens=192, repetition penalty=1.2, no-repeat 4-gram blocking=true2026.06 | 40 | 44.2 | 36.5 |