You are very much ON TIME
Record a decent little poem that I saw~
Reading Notes of ForesightKV: Optimizing KV Cache Eviction for Reasoning Models by Learning Long-Term Contribution
Something similar with the KVP (and arxived at similar time). Use SFT to train on “oracle” tokens related to future attention. Use RL/GRPO to penalize loss spikes on low-entropy (tokens that should be
Reading Notes of RLKV: Which Heads Matter for Reasoning? RL-Guided KV Cache Compression
TL;DR: Some of heads should not do KV cache eviction. This paper predicts which heads to do KV cache eviction and which heads to do full KV cache. MDP process trained with RL/GRPO based on task succes
Reading Notes of DiffAdapt: Difficulty-Adaptive Reasoning for Token-Efficient LLM Inference
Last layer hidden state already has information related to uncertainty.
Reading Notes of SWE-Pruner: Self-Adaptive Context Pruning for Coding Agents
Train a smaller model to predict what lines to be deleted for the read action of coding agent. The pruned part is sent back to the agent. The training pipeline is not clear.
Reading Notes of SideQuest: Model-Driven KV Cache Management for Long-Horizon Agentic Reasoning
Train an agent to spawn an auxiliary agent from the ongoing trajectory to select blocks of KV cache from history and delete at regular intervals.
Simulation Modeling
My class notes on simulation modeling, based on the textbook Simulation Modeling and Analysis (Averill M. Law, 5th edition, McGraw-Hill).
1. Discrete Event Simulation
What a discrete-event simulation consists of — entities, events, the event list, the clock, and statistical counters — walked through on a single-server FCFS queue.