Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domains — mentioned by nantersand

Sun, 08 Jun 2025 08:12:09 GMT

They cite https://arxiv.org/abs/2503.23829 which is interesting though. If you have lots of tokens to burn, just try the task lots of times. Can find better solutions than reasoning would on its first try. Only tested on small models though.

Paper and discussion history. A mention is not an endorsement.

Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domains — HN submission

Tue, 08 Apr 2025 17:14:10 GMT

Can reinforcement learning for LLMs scale beyond math and coding tasks? Probably

6 points; 4 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domains · HackerPapers discussions

Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domains — mentioned by nantersand

Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domains — HN submission