Information-Gain Rewards over Diversity-Pruned Tests: GT-Anchored Verifier Co-Training for Reliable Code Generation
arXiv:2609.21208v1 Announce Type: new Abstract: Self-play methods that co-train a single language model as both coder and test author promise to move code-generation RL beyond fixed test suites, but they suffer from two coupled pathologies: permissiveness collapse, where pass-rate rewards are maximised by trivial, non-discriminative tests, and concentration bias, where i.i.d. sampled tests…
Read original article on cs.AI updates on arXiv.org →