[{"date":"Oct 13, 2026","permalink":"https://www.eigenform.ai/recursive-self-improvement/continuous-learning-tool-tips/continuous-learning-llm/","summary":" A frontier model ships fixed snapshots; a local agent can learn from its environment\u0026rsquo;s …","tags":["AI","Continuous Learning","Reinforcement Learning","Fine-Tuning","LLM"],"title":"Why Doesn't DeepSeek Train Continuously? What Its Own Papers Suggest"},{"date":"Oct 09, 2026","permalink":"https://www.eigenform.ai/recursive-self-improvement/continuous-learning-tool-tips/continual-learning-with-nightly-loras/","summary":" Each night\u0026rsquo;s adapter learns from the day\u0026rsquo;s successes, so the agent can solve more the …","tags":["AI","Continuous Learning","Fine-Tuning","AI Agents"],"title":"Continuous Learning for Research via Nightly LoRAs: It's Hacky but Good"},{"date":"Oct 09, 2026","permalink":"https://www.eigenform.ai/recursive-self-improvement/continuous-learning-tool-tips/sft-vs-rl-sparse-feedback/","summary":" Match the method to the signal you have\n","tags":["AI","Continuous Learning","Fine-Tuning","Reinforcement Learning"],"title":"Can Supervised Fine-Tuning Do Something Reinforcement Learning Can't?"},{"date":"Oct 08, 2026","permalink":"https://www.eigenform.ai/recursive-self-improvement/continuous-learning-tool-tips/catastrophic-forgetting-continual-fine-tuning/","summary":" Each round of fine-tuning can erase older skills\n","tags":["AI","Continuous Learning","Fine-Tuning"],"title":"Why Can't I Just Continue SFT Forever? Catastrophic Forgetting and Ways Around It"},{"date":"Oct 08, 2026","permalink":"https://www.eigenform.ai/recursive-self-improvement/continuous-learning-tool-tips/ai-agent-memory-is-not-learning/","summary":" Memory carries facts; only weight updates change habits\n","tags":["AI","Continuous Learning","AI Agents","Fine-Tuning"],"title":"Your AI Agent Solved the Problem. Why Didn't It Learn Anything?"}]