9 citations · 26 across the 20 of their papers we have counts for
Showing 2026Show all
3 papers · 1 filter
cs.AR2026
One Distribution, Two Channels: Coverage-Calibrated KV Retention and Compensation
Shuxin Liu, Qing Liu, Yi Du +1
Long-context decoding is increasingly constrained by key--value (KV) cache memory and bandwidth. Existing fixed-budget compression methods typically separate retention from compens…
cs.LG2026
HiLoRe: What to Store, Compress, or Recompute for Efficient GRPO Training
Xinrui Chen, Mengyang Li, Ou Wu +1
Group-relative policy optimization (GRPO) makes learner-side activations a major memory-computation bottleneck: gradient checkpointing reduces activation memory through recomputati…
cs.LG2026
Token Utility Is Selection-Conditioned: Coupled Selection of Prompt Context and Response Supervision for Instruction Tuning
Can Wu, Xinrui Chen, Ou Wu +1
Token selection for instruction tuning is coupled: supervision utility depends on retained context, and vice versa. Existing methods largely optimize prompt compression or response…