Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
2
0
3
Context compaction in agentic RL can run up to 5x faster simply by streaming the KV cache instead of flushing it—accidentally turning standard LLMs into recurrent agents that preserve evicted context purely through RL.
This work presents a systematic empirical investigation of how randomness quality affects differential privacy mechanisms using IBM's DiffPrivLib, and demonstrates that DP mechanisms reliably detect deviations when approximately 1 bit in every 8 to 16 is manipulated.