← back to paper

Review history

arxiv: 2608.03573 · 2 revisions

SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs

1. 2026-08-08 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 4.0
   104751 ms 43218 in 10652 out 2026-08-08T00:44:43.032817+00:00
2. 2026-08-05 CONDITIONAL MODERATE v1.4.0-daily-deepseek-medium novelty 5.0
   177370 ms 42275 in 16282 out 2026-08-05T16:42:11.984909+00:00