You are very much ON TIME
Record a decent little poem that I saw~
Reading Notes of KVP: Learning to Evict from Key-Value Cache
Instead of evicting KV cache with checking whether previous tokens contribute to the current one, learn to predict whether past tokens contribute to the future tokens. Prediction returns a sorted list
Reading Notes of H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models
Given current token, use the query to compute attention with keys in previous tokens. Evict n tokens with least attention scores. Any token that is not relevant to current position is removed (but nev
Reading Notes of EvoDS: Self-Evolving Autonomous Data Science Agent with Skill Learning and Context Management
A data science agent (Qwen3-8B backbone) trained end-to-end to (1) synthesize and reuse new skills and (2) decide when to compress its own context, via SFT distillation + multi-role GRPO.
Reading Notes of ACM: Agentic Context Management for Long Horizon Tasks
I like the idea of insertion / deletion decisions of context management by on policy distillation, but there should be more analysis here. Also, the comparison is not fair. The compute is not saved. T
Reading Notes of ACON: Optimizing Context Compression for Long-horizon LLM Agents
I like the story of this paper to optimize cost, but it spends a long time introducing cost optimization objective while it actually uses iterative prompt engineering, which seems to be totally irrele
Blackjack 数学推导
去拉斯维加斯接触到这个东西, 查了下数学还挺有意思的 (没那么难也没那么简单那种), 一开始凭借直觉输了 70 USD, 于是回头推导了一下. 截图来自: 247blackjack.com.
