Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
Affiliation:
4
0
7
5
Training an agent to learn from an oracle's future actions, not just its initial decision, unlocks significantly deeper and more robust long-horizon reinforcement learning.
Rather than relying on scarce human demonstrations to warm up reasoning agents, dynamically scaffolding the environment's feedback allows LLMs to autonomously crack sparse-reward, long-horizon tasks and bake exploration behaviors directly into their weights.
Routing-guided exploration boosts software fix resolution rates by nearly 8%, outperforming traditional sampling methods without needing explicit answer forms.
Coding agents can now be evaluated on their ability to navigate fuzzy requirements and interactive workflows, reflecting real-world software development challenges.