single mention · post-training
Test-difficulty-driven code reward
A test-difficulty-driven code reward is introduced to mitigate the sparse reward issue for challenging code problems.1 It assigns fine-grained scores for test cases with varying difficulty levels, allowing the policy to be optimized via a dense reward signal.2
How this definition was written
Summarised by glm-5.3-flash from 1 source; every sentence is checked against the quote it cites.
To mitigate the sparse reward issue for challenging code problems, we introduce a test difficulty driven code reward
— Install the latest SGlang from main branchBy assigning fine-grained scores for test cases with varying difficulty levels, the policy can be more effectively optimized via dense reward signal
— Install the latest SGlang from main branch
Filed under post-training › reward modelling