[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tag-long-context":3},{"tag":4,"articles":11},{"id":5,"name":6,"slug":7,"article_count":8,"description_zh":9,"description_en":10},"9bacdf64-f75d-4c59-a836-d469cbe34dfc","long context","long-context",8,"長上下文指的是模型在一次推理中維持大量前後文的能力，牽涉記憶壓縮、檢索、快權重更新與推理穩定性。從 1M\u002F2M token 視窗到 state-space、TTT 與 agent 工作流，都是它的實作重點。","Long context refers to an LLM’s ability to keep and use very large histories in one pass, shaping memory design, retrieval, fast-weight updates, and stable reasoning. It shows up in 1M-2M token windows, state-space memory, TTT, and agent workflows.",[12,21,29,36,43,50],{"id":13,"slug":14,"title":15,"summary":16,"category":17,"image_url":18,"cover_image":18,"language":19,"created_at":20},"6c57f6bf-1023-4a22-a6c0-013bd88ac3d1","minimax-m1-open-hybrid-attention-reasoning-model-en","MiniMax-M1 brings 1M-token open reasoning model","MiniMax released M1, an open-source reasoning model with 1M-token context, 80k output, and low-cost API pricing.","model-release","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778797872005-z8uk.png","en","2026-05-14T22:30:39.599473+00:00",{"id":22,"slug":23,"title":24,"summary":25,"category":26,"image_url":27,"cover_image":27,"language":19,"created_at":28},"c1aac50e-0c41-471c-946e-329652f04565","sessa-attention-inside-state-space-memory-en","Sessa: Attention and State-Space Memory for Long Context","Sessa mixes attention with recurrent state-space feedback to improve long-context recall, with power-law memory tails and strong benchmark results.","research","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1776751621598-1d0l.png","2026-04-21T06:06:37.564074+00:00",{"id":30,"slug":31,"title":32,"summary":33,"category":26,"image_url":34,"cover_image":34,"language":19,"created_at":35},"b65aeb57-d1b7-4cdd-adb9-464b8cfbfe0a","in-place-ttt-llms-adapt-at-inference-en","In-Place TTT Lets LLMs Adapt at Inference","A new test-time training setup lets LLMs update fast weights in place, aiming for better long-context adaptation without full retraining.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775628426736-0dm6.png","2026-04-08T06:06:33.249426+00:00",{"id":37,"slug":38,"title":39,"summary":40,"category":17,"image_url":41,"cover_image":41,"language":19,"created_at":42},"c0e85793-59d6-47ba-9c97-f856a4544baf","grok-420-xai-flagship-model-explained-en","Grok 4.20: xAI's new flagship model explained","xAI’s Grok 4.20 adds a 2M-token context window, multi-agent reasoning, and API pricing from $2 per million input tokens.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775175184959-ok8i.png","2026-04-03T00:12:38.289208+00:00",{"id":44,"slug":45,"title":46,"summary":47,"category":17,"image_url":48,"cover_image":48,"language":19,"created_at":49},"04e78fe1-7f49-40db-bfb2-7bb4b3579276","gemini-3-1-pro-googles-top-model-in-numbers-en","Gemini 3.1 Pro: Google’s new top model in numbers","Gemini 3.1 Pro posts 77.1% on ARC-AGI-2, 94.3% on GPQA Diamond, and a 1M-token context window, while keeping Gemini 3 pricing.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775153582956-qese.png","2026-04-02T18:12:42.161483+00:00",{"id":51,"slug":52,"title":53,"summary":54,"category":26,"image_url":55,"cover_image":55,"language":19,"created_at":56},"27f0d044-b9f9-4a58-99e8-1a181ea32f19","universal-yoco-efficient-depth-scaling-en","Universal YOCO aims to scale depth without cache bloat","YOCO-U mixes recursive computation with efficient attention to scale LLM depth while keeping inference overhead and KV cache growth in check.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775115621645-wqql.png","2026-04-02T06:06:26.960639+00:00"]