Search papers, labs, and topics across Lattice.
4
0
6
3
Leveraging its own failure history, DiagEvo transforms self-play by dynamically guiding question generation to address specific reasoning weaknesses, leading to unprecedented accuracy gains.
Existing document parsers may score high on benchmarks, but they still falter on real-world tables, with a top parser achieving only 85.03 TEDS.
Combining RLVR and OPD through SAF not only prevents entropy collapse but also boosts performance across multiple benchmarks, revealing the potential for more effective reinforcement learning strategies.
Forget bigger models: massive gains in document parsing accuracy are still possible through smarter data engineering alone.
