Track trials, findings, and clinical evidence as it lands.
Search papers, labs, and topics across Lattice.
7
0
8
3
Exploiting local parameter perturbations of a teacher model unlocks complementary supervision signals, boosting LLM mathematical reasoning accuracy beyond standard on-policy self-distillation.
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.
FaithEyes reveals that self-judging mechanisms in VLMs can drastically improve tool use fidelity, leading to more reliable multimodal reasoning.
MemTrain reveals that self-supervised memory training can outperform traditional reinforcement learning approaches in enhancing LLMs' reasoning capabilities.
TrOPD stabilizes on-policy distillation by ensuring reliable teacher supervision, leading to consistent performance improvements over existing methods.
Stop cobbling together memory-augmented agents: MemFactory offers a unified "Lego-like" framework that streamlines training and boosts performance by up to 14.8%.
Pointwise reward models can finally compete with pairwise models in RLHF, thanks to a new intergroup comparison method that scales linearly with the number of candidates.