Recovering Reliable Supervision via Critique-Guided On-Policy Distillation

Researchers introduce Latent Compaction, a test-time training method that compresses KV caches 32× while maintaining 96.7% accuracy on retrieval tasks.

Researchers prove first-price auctions in AI agent markets can achieve near-optimal welfare under competitive floors and conditional independence assumptions.

Researchers introduce a retraining-free method to make transformers elastic, enabling on-the-fly adaptation across computational budgets by jointly ranking MLP neurons and attention heads.

Researchers propose Co-Evolving Acting and Forecasting (name), a method that improves language agent training efficiency by adding multi-step action targets to standard RL approaches.

Researchers propose HiRe-MoMa, a hierarchical residual learning framework that enables real-world robot fine-tuning with only 30 minutes of human-guided interaction.